<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "http://jats.nlm.nih.gov/publishing/1.3/JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xml:lang="en" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">
<processing-meta>
<custom-meta-group content-type="composition">
<custom-meta specific-use="newgen" xlink:href="https://www.newgen.co/">
<meta-name>Composition Vendor</meta-name>
<meta-value>Newgen KnowledgeWorks (P) Ltd.</meta-value>
</custom-meta>
</custom-meta-group>
</processing-meta>
<front>
<journal-meta>
<journal-id journal-id-type="nlm-ta">PLoS One</journal-id>
<journal-id journal-id-type="publisher-id">plos</journal-id>
<journal-id journal-id-type="pmc">plosone</journal-id>
<journal-title-group>
<journal-title>PLOS One</journal-title>
</journal-title-group>
<issn pub-type="epub">1932-6203</issn>
<publisher>
<publisher-name>Public Library of Science</publisher-name>
<publisher-loc>San Francisco, CA USA</publisher-loc>
</publisher>
</journal-meta>
<article-meta>
<article-id pub-id-type="doi">10.1371/journal.pone.0343751</article-id>
<article-id pub-id-type="publisher-id">PONE-D-26-05935</article-id>
<article-categories>
<subj-group subj-group-type="heading">
<subject>Research Article</subject>
</subj-group>
<subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Oncology</subject><subj-group><subject>Cancers and neoplasms</subject><subj-group><subject>Genitourinary tract tumors</subject><subj-group><subject>Prostate cancer</subject></subj-group></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Urology</subject><subj-group><subject>Prostate diseases</subject><subj-group><subject>Prostate cancer</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Diagnostic medicine</subject><subj-group><subject>Cancer detection and diagnosis</subject></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Oncology</subject><subj-group><subject>Cancer detection and diagnosis</subject></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Surgical and invasive medical procedures</subject><subj-group><subject>Biopsy</subject></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Diagnostic medicine</subject><subj-group><subject>Cancer detection and diagnosis</subject><subj-group><subject>Cancer screening</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Oncology</subject><subj-group><subject>Cancer detection and diagnosis</subject><subj-group><subject>Cancer screening</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Epidemiology</subject><subj-group><subject>Medical risk factors</subject><subj-group><subject>Cancer risk factors</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Oncology</subject><subj-group><subject>Cancer risk factors</subject></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Diagnostic medicine</subject><subj-group><subject>Diagnostic radiology</subject><subj-group><subject>Magnetic resonance imaging</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Research and analysis methods</subject><subj-group><subject>Imaging techniques</subject><subj-group><subject>Diagnostic radiology</subject><subj-group><subject>Magnetic resonance imaging</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Radiology and imaging</subject><subj-group><subject>Diagnostic radiology</subject><subj-group><subject>Magnetic resonance imaging</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Medicine and health sciences</subject><subj-group><subject>Urology</subject><subj-group><subject>Genitourinary infections</subject><subj-group><subject>Urinary tract infections</subject></subj-group></subj-group></subj-group></subj-group><subj-group subj-group-type="Discipline-v3">
<subject>Research and analysis methods</subject><subj-group><subject>Mathematical and statistical techniques</subject><subj-group><subject>Mathematical functions</subject><subj-group><subject>Curve fitting</subject></subj-group></subj-group></subj-group></subj-group></article-categories>
<title-group>
<article-title>Joint modelling of PSA dynamics and prostate cancer risks: A population-based study in men without prior prostate cancer</article-title>
<alt-title alt-title-type="running-head">Joint modelling of PSA dynamics and prostate cancer risks</alt-title>
</title-group>
<contrib-group>
<contrib contrib-type="author" corresp="yes" equal-contrib="yes" xlink:type="simple">
<contrib-id authenticated="true" contrib-id-type="orcid">https://orcid.org/0009-0007-2327-6156</contrib-id>
<name name-style="western">
<surname>Akynkozhayev</surname>
<given-names>Birzhan</given-names>
</name>
<role content-type="http://credit.niso.org/contributor-roles/conceptualization/">Conceptualization</role>
<role content-type="http://credit.niso.org/contributor-roles/data-curation/">Data curation</role>
<role content-type="http://credit.niso.org/contributor-roles/formal-analysis/">Formal analysis</role>
<role content-type="http://credit.niso.org/contributor-roles/investigation/">Investigation</role>
<role content-type="http://credit.niso.org/contributor-roles/methodology/">Methodology</role>
<role content-type="http://credit.niso.org/contributor-roles/software/">Software</role>
<role content-type="http://credit.niso.org/contributor-roles/validation/">Validation</role>
<role content-type="http://credit.niso.org/contributor-roles/visualization/">Visualization</role>
<role content-type="http://credit.niso.org/contributor-roles/writing-original-draft/">Writing – original draft</role>
<role content-type="http://credit.niso.org/contributor-roles/writing-review-editing/">Writing – review &amp; editing</role>
<xref ref-type="aff" rid="aff001"><sup>1</sup></xref>
<xref ref-type="corresp" rid="cor001">*</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple">
<name name-style="western">
<surname>Christoffersen</surname>
<given-names>Benjamin</given-names>
</name>
<role content-type="http://credit.niso.org/contributor-roles/conceptualization/">Conceptualization</role>
<role content-type="http://credit.niso.org/contributor-roles/methodology/">Methodology</role>
<role content-type="http://credit.niso.org/contributor-roles/software/">Software</role>
<role content-type="http://credit.niso.org/contributor-roles/supervision/">Supervision</role>
<role content-type="http://credit.niso.org/contributor-roles/validation/">Validation</role>
<role content-type="http://credit.niso.org/contributor-roles/writing-review-editing/">Writing – review &amp; editing</role>
<xref ref-type="aff" rid="aff001"><sup>1</sup></xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple">
<name name-style="western">
<surname>Lantz</surname>
<given-names>Anna</given-names>
</name>
<role content-type="http://credit.niso.org/contributor-roles/writing-review-editing/">Writing – review &amp; editing</role>
<xref ref-type="aff" rid="aff001"><sup>1</sup></xref>
<xref ref-type="aff" rid="aff002"><sup>2</sup></xref>
<xref ref-type="fn" rid="econtrib001"><sup>‡</sup></xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple">
<name name-style="western">
<surname>Nordström</surname>
<given-names>Tobias</given-names>
</name>
<role content-type="http://credit.niso.org/contributor-roles/writing-review-editing/">Writing – review &amp; editing</role>
<xref ref-type="aff" rid="aff001"><sup>1</sup></xref>
<xref ref-type="aff" rid="aff003"><sup>3</sup></xref>
<xref ref-type="fn" rid="econtrib001"><sup>‡</sup></xref>
</contrib>
<contrib contrib-type="author" equal-contrib="yes" xlink:type="simple">
<name name-style="western">
<surname>Humphreys</surname>
<given-names>Keith</given-names>
</name>
<role content-type="http://credit.niso.org/contributor-roles/supervision/">Supervision</role>
<role content-type="http://credit.niso.org/contributor-roles/validation/">Validation</role>
<role content-type="http://credit.niso.org/contributor-roles/writing-review-editing/">Writing – review &amp; editing</role>
<xref ref-type="aff" rid="aff001"><sup>1</sup></xref>
</contrib>
<contrib contrib-type="author" equal-contrib="yes" xlink:type="simple">
<name name-style="western">
<surname>Clements</surname>
<given-names>Mark</given-names>
</name>
<role content-type="http://credit.niso.org/contributor-roles/conceptualization/">Conceptualization</role>
<role content-type="http://credit.niso.org/contributor-roles/data-curation/">Data curation</role>
<role content-type="http://credit.niso.org/contributor-roles/funding-acquisition/">Funding acquisition</role>
<role content-type="http://credit.niso.org/contributor-roles/methodology/">Methodology</role>
<role content-type="http://credit.niso.org/contributor-roles/project-administration/">Project administration</role>
<role content-type="http://credit.niso.org/contributor-roles/resources/">Resources</role>
<role content-type="http://credit.niso.org/contributor-roles/supervision/">Supervision</role>
<role content-type="http://credit.niso.org/contributor-roles/validation/">Validation</role>
<role content-type="http://credit.niso.org/contributor-roles/writing-original-draft/">Writing – original draft</role>
<role content-type="http://credit.niso.org/contributor-roles/writing-review-editing/">Writing – review &amp; editing</role>
<xref ref-type="aff" rid="aff001"><sup>1</sup></xref>
</contrib>
</contrib-group>
<aff id="aff001"><label>1</label> <addr-line>Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden</addr-line></aff>
<aff id="aff002"><label>2</label> <addr-line>Department of Urology, Karolinska University Hospital Solna, Stockholm, Sweden</addr-line></aff>
<aff id="aff003"><label>3</label> <addr-line>Department of Clinical Sciences, Danderyds Sjukhus, Stockholm, Sweden</addr-line></aff>
<contrib-group>
<contrib contrib-type="editor" xlink:type="simple">
<name name-style="western">
<surname>Wroński</surname>
<given-names>Stanisław Jacek</given-names>
</name>
<role>Editor</role>
<xref ref-type="aff" rid="edit1"/></contrib>
</contrib-group>
<aff id="edit1"><addr-line>Jan Biziel University Hospital No 2 in Bydgoszcz: Szpital Uniwersytecki Nr 2 im dr Jana Biziela w Bydgoszczy, POLAND</addr-line></aff>
<author-notes>
<corresp id="cor001">* E-mail: <email xlink:type="simple">birzhan.akynkozhayev@ki.se</email></corresp>
<fn fn-type="conflict" id="coi001">
<p>I have read the journal’s policy and the authors of this manuscript have the following competing interests: Dr Nordström reported owning shares in A3P Biomedical. No other disclosures were reported. This does not alter our adherence to PLOS ONE policies on sharing data and materials.</p>
</fn>
<fn fn-type="other" id="econtrib001">
<p>‡ These authors also contributed equally to this work.</p>
</fn>
</author-notes>
<pub-date pub-type="epub"><day>10</day><month>7</month><year>2026</year></pub-date>
<pub-date pub-type="collection"><year>2026</year></pub-date>
<volume>21</volume>
<issue>7</issue>
<elocation-id>e0343751</elocation-id>
<history>
<date date-type="received"><day>10</day><month>2</month><year>2026</year></date>
<date date-type="accepted"><day>27</day><month>6</month><year>2026</year></date>
</history>
<permissions>
<copyright-year>2026</copyright-year>
<copyright-holder>Akynkozhayev et al</copyright-holder>
<license xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">
<license-p>This is an open access article distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">Creative Commons Attribution License</ext-link>, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.</license-p></license>
</permissions>
<self-uri content-type="pdf" xlink:href="info:doi/10.1371/journal.pone.0343751"/>
<abstract>
<p>While the prostate-specific antigen (PSA) test is a widely used prostate cancer screening tool, its application remains controversial. Opportunistic PSA testing generates complex data in which testing intensities, PSA levels, and prostate cancer diagnosis are interdependent. Conventional analyses rarely model these processes jointly. The objective of this study was to develop a population-based joint model to analyse PSA dynamics, retesting patterns, and prostate cancer risk. We used the Stockholm Prostate Cancer Diagnostics Register to identify 506,761 men with at least one PSA test between 2003 and 2020. We fitted a joint model linking three components: a linear mixed-effects submodel for PSA over age, and two proportional hazards submodels for time to next PSA test and time to prostate cancer diagnosis. PSA increased nonlinearly with age, with substantial between-person heterogeneity and increasing unexplained variation with increasing age. Estimates from models fitted to each time-to-event process separately were attenuated relative to the joint model: hazard ratios per doubling of PSA were 1.61 (95% CI: 1.59-1.62; P &lt; .001) versus 2.01 (95% CI: 1.99-2.02; P &lt; .001) for prostate cancer diagnosis, respectively, and 1.068 (95% CI: 1.066-1.070; P &lt; .001) versus 1.163 (95% CI: 1.161-1.165; P &lt; .001) for retesting, respectively. This pattern is consistent with informative observation bias: men with higher PSA values are tested more frequently, so that part of the association between PSA and the event is absorbed by the unmodelled observation mechanism when the processes are analysed separately. Jointly modelling PSA, testing intensity, and diagnosis through shared random effects accounts for this dependence. As a limitation, the study is primarily limited by its observational nature and a lack of data on non-cancer factors that can elevate PSA, such as urinary tract infections or lower urinary tract symptoms. In conclusion, jointly modelling PSA dynamics and testing behaviour corrects for the informative observation bias inherent in opportunistic testing. This approach yields more accurate population estimates compared to traditional isolated models. Our findings suggest that PSA dynamics may be clinically informative and that screening models should jointly incorporate testing history and PSA trajectories to improve precision.</p>
</abstract>
<funding-group>
<award-group id="award001">
<funding-source>
<institution-wrap>
<institution-id institution-id-type="funder-id">http://dx.doi.org/10.13039/501100004359</institution-id>
<institution>Vetenskapsrådet</institution>
</institution-wrap>
</funding-source><award-id>2025-03045_VR</award-id>
<principal-award-recipient><name name-style="western">
<surname>Clements</surname><given-names>Mark</given-names></name></principal-award-recipient></award-group>
<award-group id="award002">
<funding-source>
<institution-wrap>
<institution-id institution-id-type="funder-id">http://dx.doi.org/10.13039/501100004359</institution-id>
<institution>Vetenskapsrådet</institution>
</institution-wrap>
</funding-source><award-id>environment grant 2024-06502</award-id>
<principal-award-recipient><name name-style="western">
<surname>Clements</surname><given-names>Mark</given-names></name></principal-award-recipient></award-group>
<award-group id="award003">
<funding-source>
<institution-wrap>
<institution-id institution-id-type="funder-id">http://dx.doi.org/10.13039/100017156</institution-id>
<institution>Swedish e-Science Research Centre</institution>
</institution-wrap>
</funding-source><principal-award-recipient><name name-style="western">
<surname>Clements</surname><given-names>Mark</given-names></name></principal-award-recipient></award-group>
<award-group id="award004">
<funding-source>
<institution-wrap>
<institution-id institution-id-type="funder-id">http://dx.doi.org/10.13039/501100002794</institution-id>
<institution>Cancerfonden</institution>
</institution-wrap>
</funding-source><award-id>CAN21/1512</award-id>
<principal-award-recipient><name name-style="western">
<surname>Clements</surname><given-names>Mark</given-names></name></principal-award-recipient></award-group>
<award-group id="award005">
<funding-source><institution>Swedish Prostate Cancer Society</institution>
</funding-source><principal-award-recipient><name name-style="western">
<surname>Clements</surname><given-names>Mark</given-names></name></principal-award-recipient></award-group>
<award-group id="award006">
<funding-source>
<institution-wrap>
<institution-id institution-id-type="funder-id">http://dx.doi.org/10.13039/100010663</institution-id>
<institution>H2020 European Research Council</institution>
</institution-wrap>
</funding-source><award-id>HEAP grant No. 874662</award-id>
<principal-award-recipient><name name-style="western">
<surname>Clements</surname><given-names>Mark</given-names></name></principal-award-recipient></award-group>
<funding-statement>This work was supported by the Swedish Research Council (Vetenskapsrådet) (grant 2025-03045 to MC; environment grant 2024-06502 to MC); the Swedish e-Science Research Centre (SeRC) (to MC); the Swedish Cancer Society (Cancerfonden) (grant CAN21/1512 to MC); the Swedish Prostate Cancer Society (Prostatacancerförbundet) (to MC); and the European Commission (HEAP grant No. 874662 to MC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.</funding-statement>
</funding-group>
<counts>
<fig-count count="4"/>
<table-count count="2"/>
<page-count count="12"/>
</counts>
<custom-meta-group>
<custom-meta id="data-availability">
<meta-name>Data Availability</meta-name>
<meta-value>Data cannot be shared publicly because of restrictions under the Swedish Public Access and Secrecy Act (Offentlighets- och sekretesslag 2009:400) on individual-level health register data, and because the data are owned by the STHLM0 Steering Committee at Karolinska Institutet rather than by the authors. Long-term data storage and availability are maintained by the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, which hosts the STHLM0 register. Researchers who meet the criteria for access to confidential register data may submit a request through the STHLM0 web portal (<ext-link ext-link-type="uri" xlink:href="https://ki.se/en/research/research-areas-centres-and-networks/research-groups/henrik-gronbergs-research-group#tab-sthlm0" xlink:type="simple">https://ki.se/en/research/research-areas-centres-and-networks/research-groups/henrik-gronbergs-research-group#tab-sthlm0)</ext-link>. Following approval by the steering committee, applicants must obtain ethical approval from the Swedish Ethical Review Authority (<ext-link ext-link-type="uri" xlink:href="https://www.etikprovningsansokan.se/" xlink:type="simple">https://www.etikprovningsansokan.se/)</ext-link>. Once the Swedish Ethical Review Authority approves the project, a data request form can be completed at the STHLM0 website, and applicants may contact the STHLM0 Steering Committee at Martin Eklund (<email xlink:type="simple">martin.eklund@ki.se</email>), or Henrik Grönberg (<email xlink:type="simple">henrik.gronberg@ki.se</email>).</meta-value>
</custom-meta>
</custom-meta-group>
</article-meta>
</front>
<body>
<sec id="sec001" sec-type="intro">
<title>Introduction</title>
<p>Prostate cancer is the most common cancer and is the third leading cause of cancer-related death among men in the European Union [<xref ref-type="bibr" rid="pone.0343751.ref001">1</xref>]. PSA is an inexpensive blood test that has long guided decisions on prostate cancer screening. The use of a single PSA threshold remains controversial [<xref ref-type="bibr" rid="pone.0343751.ref002">2</xref>–<xref ref-type="bibr" rid="pone.0343751.ref004">4</xref>]. The European Randomized Study of Screening for Prostate Cancer [<xref ref-type="bibr" rid="pone.0343751.ref005">5</xref>] and the Cluster Randomized Trial of PSA Testing for Prostate Cancer [<xref ref-type="bibr" rid="pone.0343751.ref006">6</xref>] demonstrated significant reductions in prostate cancer mortality, whereas the U.S. Prostate, Lung, Colorectal and Ovarian trial found no such reduction [<xref ref-type="bibr" rid="pone.0343751.ref007">7</xref>]. PSA levels increase with age and age-specific thresholds have been proposed to improve screening accuracy, yet their use remains controversial [<xref ref-type="bibr" rid="pone.0343751.ref008">8</xref>–<xref ref-type="bibr" rid="pone.0343751.ref010">10</xref>].</p>
<p>Longitudinal PSA trajectories may provide information beyond a single measurement. PSA doubling times have been linked to aggressive disease [<xref ref-type="bibr" rid="pone.0343751.ref011">11</xref>,<xref ref-type="bibr" rid="pone.0343751.ref012">12</xref>], although population-level studies have questioned their predictive value [<xref ref-type="bibr" rid="pone.0343751.ref013">13</xref>,<xref ref-type="bibr" rid="pone.0343751.ref014">14</xref>].</p>
<p>Large trials and health-system cohorts have documented extensive repeat PSA testing in routine care [<xref ref-type="bibr" rid="pone.0343751.ref015">15</xref>–<xref ref-type="bibr" rid="pone.0343751.ref020">20</xref>], but most analyses have considered only one of three key processes: longitudinal PSA trajectories, testing patterns (the observational process), or time to prostate cancer diagnosis. Few studies have applied joint modelling frameworks, which link longitudinal measurements with time-to-event outcomes, to PSA data, but these have focused on men after a prostate cancer diagnosis [<xref ref-type="bibr" rid="pone.0343751.ref021">21</xref>–<xref ref-type="bibr" rid="pone.0343751.ref023">23</xref>]. There is limited evidence on how PSA trajectories, repeat testing, and diagnosis are inter-related in men preceding any diagnosis of prostate cancer. Moreover, existing studies have not explicitly accounted for the fact that men with higher PSA values are more likely to undergo frequent testing. Recent computational developments now make a joint analysis of all three processes feasible for larger register-based datasets.</p>
<p>In this study, we develop a detailed population-based PSA model for men without a prior prostate cancer diagnosis. Using extensive PSA test data linked with rich health and population registers, we describe: (a) PSA trajectories, which are allowed to vary between individuals; (b) time to prostate cancer diagnosis; and (c) PSA testing patterns, with both diagnosis and testing dependent on the underlying PSA trajectory. Jointly modelling these longitudinal and time-to-event processes may provide novel insights into both the PSA dynamics and the rescreening behaviour. Specifically, we expect that estimates from the joint model will more accurately capture the underlying data-generating mechanisms than those from separate models. Patient-level posterior estimates of the random effects from such a model could serve as a basis for identifying men whose PSA levels are rising faster than expected of their age, signalling the need for closer clinical follow-up.</p>
</sec>
<sec id="sec002" sec-type="materials|methods">
<title>Materials and methods</title>
<sec id="sec003">
<title>Materials</title>
<p>This study uses data from the Stockholm Prostate Cancer Diagnostics Register [<xref ref-type="bibr" rid="pone.0343751.ref024">24</xref>], a comprehensive regional database that links information from Swedish national and regional registers, including PSA test dates and results, biopsies, cancer diagnoses, cause of death, and population data for men residing in Stockholm County. These data reflect a healthcare system where organised prostate cancer screening has not been recommended by the Swedish Board of Health and Welfare. Men may instead discuss prostate cancer testing with their clinician through a process of shared decision-making. Diagnostic practice for men who chose to have a PSA test evolved during the study window. In the earlier years of the cohort (approximately 2003–2015), an elevated PSA was typically followed by a systematic transrectal ultrasound-guided biopsy. Pre-biopsy magnetic resonance imaging (MRI) in men with elevated PSA has been recommended by international guidelines since 2015, and this recommendation was incorporated into the Swedish national guidelines (Nationellt vårdprogram för prostatacancer) in 2020 [<xref ref-type="bibr" rid="pone.0343751.ref025">25</xref>]. The contemporary pathway in Stockholm is therefore PSA-MRI-biopsy in selected men with an elevated PSA. The specific age-dependent PSA thresholds for referral into a standardised care pathway (Standardiserat vårdförlopp) are ≥ 3.0 ng/mL for men under 70 years, ≥ 5.0 ng/mL for those aged 70–80, and ≥ 7.0 ng/mL for those over 80. In clinical practice, if an initial PSA result is between 3 and 10 ng/mL and the prostate palpation is benign, a second PSA test is recommended after approximately three weeks to account for intra-individual variation before proceeding to MRI. Further, the indications for biopsy are refined using PSA density; for example, a biopsy is recommended for PI-RADS 3 findings only if the PSA density is ≥ 0.10 ng/mL/cm<sup>3</sup>, or for PI-RADS 1–2 findings if the density is ≥ 0.20 ng/mL/cm<sup>3</sup>.</p>
<p>The data were accessed on 12 January 2023 under the ethical approval from the Regional Ethical Review Board in Stockholm (Dnr 2012/438–31/3), with a subsequent amendment (Dnr 2016/620–32). As this was a register-based study, the requirement for individual informed consent was waived by the ethics committee. All data were processed anonymously, and the authors had no access to information that could be used to identify individual participants during or after the study.</p>
<p>The study population comprises 506,761 men who underwent at least one PSA test between 1 January 2003 and 31 December 2020. Disease verification was register-based and relied on biopsy-confirmed diagnoses, with predominantly systematic biopsies before 2018 and a rapidly increasing use of pre-biopsy MRI thereafter. PSA values and testing patterns change drastically after diagnosis and treatment, and the post-diagnosis is not the focus of this study; we therefore censored follow-up at the date of prostate cancer diagnosis and excluded all tests taken after the diagnosis date.</p>
</sec>
<sec id="sec004">
<title>Statistical methods</title>
<p>In outline, we fitted a joint model that linked three processes: the longitudinal trajectory of log-transformed PSA over age, time to prostate cancer diagnosis, and time to PSA test. The longitudinal submodel used a linear mixed-effects model to describe individual log-transformed PSA values as a nonlinear function of attained age, represented using a natural cubic spline with three degrees of freedom. Patient-level random intercepts and slopes were included to allow individual log-transformed PSA trajectories to deviate from the population-average trend.</p>
<p>Time to prostate cancer diagnosis was modelled using a left-truncated proportional hazards model, with the hazard proportional to: (i) a baseline hazard specified as a function of attained age using a natural cubic spline with three degrees of freedom; and to (ii) the underlying log-transformed PSA trajectory defined in the longitudinal submodel.</p>
<p>Time to PSA testing was modelled using a recurrent-event proportional hazards model in a counting-process formulation, with the hazard proportional to: (i) a baseline hazard specified as a function of attained age using a natural cubic spline with three degrees of freedom; and to (ii) the underlying log-transformed PSA trajectory from the longitudinal submodel. This formulation accounts for the correlation among repeated tests within individuals.</p>
<p>Correlations between the three processes were thus modelled through the shared underlying longitudinal PSA trajectory, which was modelled explicitly as a function of attained age. Each submodel can therefore be viewed implicitly as a model of attained age. Additionally, from the fitted joint model, we were able to estimate patient-specific trajectories for all three processes.</p>
<p>More formally, let individual <inline-formula id="pone.0343751.e001"><alternatives><graphic id="pone.0343751.e001g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e001" xlink:type="simple"/><mml:math display="inline" id="M1"><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mo>=</mml:mo><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow><mml:mo>,</mml:mo><mml:mi>…</mml:mi><mml:mo>,</mml:mo><mml:mrow><mml:mi>n</mml:mi></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> have <inline-formula id="pone.0343751.e002"><alternatives><graphic id="pone.0343751.e002g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e002" xlink:type="simple"/><mml:math display="inline" id="M2"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> PSA tests during the follow-up, with <inline-formula id="pone.0343751.e003"><alternatives><graphic id="pone.0343751.e003g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e003" xlink:type="simple"/><mml:math display="inline" id="M3"><mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula>-th test (<inline-formula id="pone.0343751.e004"><alternatives><graphic id="pone.0343751.e004g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e004" xlink:type="simple"/><mml:math display="inline" id="M4"><mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow><mml:mo>=</mml:mo><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow><mml:mo>,</mml:mo><mml:mi>…</mml:mi><mml:mo>,</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula>) taken at age <inline-formula id="pone.0343751.e005"><alternatives><graphic id="pone.0343751.e005g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e005" xlink:type="simple"/><mml:math display="inline" id="M5"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula>. Denote the corresponding log PSA value by <inline-formula id="pone.0343751.e006"><alternatives><graphic id="pone.0343751.e006g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e006" xlink:type="simple"/><mml:math display="inline" id="M6"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>Y</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula>. The longitudinal submodel was</p>
<disp-formula id="pone.0343751.e007"><alternatives><graphic id="pone.0343751.e007g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e007" xlink:type="simple"/><mml:math display="block" id="M7"><mml:mrow><mml:mtable><mml:mtr><mml:mtd><mml:msub><mml:mrow><mml:mrow><mml:mi>Y</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo>=</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">X</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">β</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">Z</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">b</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>ϵ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo>,</mml:mo></mml:mtd></mml:mtr></mml:mtable></mml:mrow></mml:math></alternatives></disp-formula>
<p>where <inline-formula id="pone.0343751.e008"><alternatives><graphic id="pone.0343751.e008g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e008" xlink:type="simple"/><mml:math display="inline" id="M8"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">X</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> was the fixed-effects design matrix containing an intercept and a natural cubic spline on age at test with two interior knots placed at the tertiles of the observed age-at-test distribution, with <inline-formula id="pone.0343751.e009"><alternatives><graphic id="pone.0343751.e009g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e009" xlink:type="simple"/><mml:math display="inline" id="M9"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">X</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> representing the row corresponding to individual <inline-formula id="pone.0343751.e010"><alternatives><graphic id="pone.0343751.e010g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e010" xlink:type="simple"/><mml:math display="inline" id="M10"><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> at his <inline-formula id="pone.0343751.e011"><alternatives><graphic id="pone.0343751.e011g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e011" xlink:type="simple"/><mml:math display="inline" id="M11"><mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula>th test, <inline-formula id="pone.0343751.e012"><alternatives><graphic id="pone.0343751.e012g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e012" xlink:type="simple"/><mml:math display="inline" id="M12"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">β</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> was the vector of fixed-effects parameters. <inline-formula id="pone.0343751.e013"><alternatives><graphic id="pone.0343751.e013g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e013" xlink:type="simple"/><mml:math display="inline" id="M13"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">Z</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> was the random-effects design matrix, containing an intercept and a random slope for age, and <inline-formula id="pone.0343751.e014"><alternatives><graphic id="pone.0343751.e014g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e014" xlink:type="simple"/><mml:math display="inline" id="M14"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">b</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> were the individual-specific random effects. We assumed <inline-formula id="pone.0343751.e015"><alternatives><graphic id="pone.0343751.e015g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e015" xlink:type="simple"/><mml:math display="inline" id="M15"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">b</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mi>~</mml:mi><mml:mrow><mml:mi>N</mml:mi></mml:mrow><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">μ</mml:mi></mml:mrow></mml:mrow><mml:mo>,</mml:mo><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">Σ</mml:mi></mml:mrow></mml:mrow><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> with mean vector <inline-formula id="pone.0343751.e016"><alternatives><graphic id="pone.0343751.e016g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e016" xlink:type="simple"/><mml:math display="inline" id="M16"><mml:mrow><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">μ</mml:mi></mml:mrow></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> and covariance matrix</p>
<p><inline-formula id="pone.0343751.e017"><alternatives><graphic id="pone.0343751.e017g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e017" xlink:type="simple"/><mml:math display="inline" id="M17"><mml:mrow><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">Σ</mml:mi></mml:mrow><mml:mo>=</mml:mo><mml:mtext> </mml:mtext></mml:mrow><mml:mo>[</mml:mo><mml:mtable><mml:mtr><mml:mtd><mml:munderover><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>0</mml:mtext></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>2</mml:mtext></mml:mrow></mml:mrow></mml:munderover></mml:mtd><mml:mtd><mml:mrow><mml:mi>ρ</mml:mi></mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>0</mml:mtext></mml:mrow></mml:mrow></mml:msub><mml:msub><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow></mml:mrow></mml:msub></mml:mtd></mml:mtr><mml:mtr><mml:mtd><mml:mrow><mml:mi>ρ</mml:mi></mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>0</mml:mtext></mml:mrow></mml:mrow></mml:msub><mml:msub><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow></mml:mrow></mml:msub></mml:mtd><mml:mtd><mml:munderover><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>2</mml:mtext></mml:mrow></mml:mrow></mml:munderover></mml:mtd></mml:mtr></mml:mtable><mml:mo>]</mml:mo></mml:mrow></mml:math></alternatives></inline-formula>,</p>
<p>where <inline-formula id="pone.0343751.e018"><alternatives><graphic id="pone.0343751.e018g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e018" xlink:type="simple"/><mml:math display="inline" id="M18"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>0</mml:mtext></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula>and <inline-formula id="pone.0343751.e019"><alternatives><graphic id="pone.0343751.e019g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e019" xlink:type="simple"/><mml:math display="inline" id="M19"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> were the standard deviations of the random intercept and random slope, respectively, and <inline-formula id="pone.0343751.e020"><alternatives><graphic id="pone.0343751.e020g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e020" xlink:type="simple"/><mml:math display="inline" id="M20"><mml:mrow><mml:mrow><mml:mi>ρ</mml:mi></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> was their correlation. <inline-formula id="pone.0343751.e021"><alternatives><graphic id="pone.0343751.e021g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e021" xlink:type="simple"/><mml:math display="inline" id="M21"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>ϵ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mi>~</mml:mi><mml:mrow><mml:mi>N</mml:mi></mml:mrow><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:mrow><mml:mtext>0</mml:mtext></mml:mrow><mml:mo>,</mml:mo><mml:msup><mml:mrow><mml:mrow><mml:mi>σ</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mtext>2</mml:mtext></mml:mrow></mml:mrow></mml:msup><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> was the independent measurement error. Disregarding the sharing of random effects with the survival components, this is a standard linear mixed-effects model.</p>
<p>The hazard functions at age <inline-formula id="pone.0343751.e022"><alternatives><graphic id="pone.0343751.e022g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e022" xlink:type="simple"/><mml:math display="inline" id="M22"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> were defined as</p>
<disp-formula id="pone.0343751.e023"><alternatives><graphic id="pone.0343751.e023g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e023" xlink:type="simple"/><mml:math display="block" id="M23"><mml:mrow><mml:mtable><mml:mtr><mml:mtd><mml:msub><mml:mrow><mml:mrow><mml:mi>h</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow><mml:mo>=</mml:mo><mml:mrow><mml:mtext>exp</mml:mtext></mml:mrow><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">X</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>s</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:msub><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">β</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>s</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>α</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">Z</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>m</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mrow><mml:mo fence="true" form="prefix" stretchy="true">(</mml:mo><mml:msub><mml:mrow><mml:mrow><mml:mi>t</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow><mml:mrow><mml:mi>j</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">b</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>i</mml:mi></mml:mrow></mml:mrow></mml:msub><mml:mo fence="true" form="postfix" stretchy="true">)</mml:mo></mml:mrow><mml:mo>,</mml:mo></mml:mtd></mml:mtr></mml:mtable></mml:mrow></mml:math></alternatives></disp-formula>
<p>where <inline-formula id="pone.0343751.e024"><alternatives><graphic id="pone.0343751.e024g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e024" xlink:type="simple"/><mml:math display="inline" id="M24"><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow><mml:mo>=</mml:mo><mml:mrow><mml:mtext>1</mml:mtext></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> corresponded to the hazard of diagnosis and <inline-formula id="pone.0343751.e025"><alternatives><graphic id="pone.0343751.e025g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e025" xlink:type="simple"/><mml:math display="inline" id="M25"><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow><mml:mo>=</mml:mo><mml:mrow><mml:mtext>2</mml:mtext></mml:mrow></mml:mrow></mml:math></alternatives></inline-formula> corresponded to the intensity of the next PSA test. <inline-formula id="pone.0343751.e026"><alternatives><graphic id="pone.0343751.e026g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e026" xlink:type="simple"/><mml:math display="inline" id="M26"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">X</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>s</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> was the design matrix for the fixed effects, specified similarly to the longitudinal submodel to include an intercept and a natural cubic spline on age with two interior knots placed at the tertiles of the observed event and censoring times. <inline-formula id="pone.0343751.e027"><alternatives><graphic id="pone.0343751.e027g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e027" xlink:type="simple"/><mml:math display="inline" id="M27"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi mathvariant="bold-italic">β</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>s</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> was the vector of fixed‐effects parameters, and <inline-formula id="pone.0343751.e028"><alternatives><graphic id="pone.0343751.e028g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e028" xlink:type="simple"/><mml:math display="inline" id="M28"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>α</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula> represented the association parameter linking the longitudinal random effects to each hazard process. The models accounted for left-truncation to adjust for the delayed entry of individuals at varying ages. Disregarding the sharing of random effects through <inline-formula id="pone.0343751.e029"><alternatives><graphic id="pone.0343751.e029g" mimetype="image" position="anchor" xlink:href="info:doi/10.1371/journal.pone.0343751.e029" xlink:type="simple"/><mml:math display="inline" id="M29"><mml:mrow><mml:msub><mml:mrow><mml:mrow><mml:mi>α</mml:mi></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></alternatives></inline-formula>, these are standard proportional hazards models with log-normal frailties.</p>
<p>The model was fitted using a frequentist marginal maximum likelihood approach. To approximate the intractable integrals over the random effects, we used Gaussian variational approximations [<xref ref-type="bibr" rid="pone.0343751.ref026">26</xref>]. Technical details of our novel implementation are provided in a separate paper [<xref ref-type="bibr" rid="pone.0343751.ref027">27</xref>]. To illustrate the fitted joint model, we presented three kinds of output: (i) the population-average log PSA trajectory as a function of age, derived from the fixed-effects part of the longitudinal submodel; (ii) hazard ratios per doubling of PSA for the diagnosis and retesting processes, obtained from the association parameters; and (iii) patient-specific estimated curves for all three processes, including the log PSA trajectory, the hazard of retesting, and the hazard of diagnosis, where each individual’s random effects were estimated from their variational posterior (analogous to best linear unbiased predictors in linear mixed-effects models) and combined with the corresponding population-average curve. For (iii), the uncertainty bands were derived from the variational covariance of the random effects.To contrast the joint model with modelling each process in isolation, we fitted corresponding standalone models to each component. Longitudinal PSA values were modelled using a linear mixed-effects model with attained age represented by a natural cubic spline with three degrees of freedom, matching the longitudinal submodel of the joint model. Time to prostate cancer diagnosis was modelled using a left-truncated proportional hazards model with attained age as the timescale, so the baseline hazard is a function of age, estimated non-parametrically. We additionally fitted a Royston-Parmar flexible parametric proportional hazards model, with the baseline log-cumulative hazard given by a natural cubic spline on age with three degrees of freedom, matching the joint model’s parameterisation. The PSA-testing process was modelled using a Prentice–Williams–Peterson (PWP) recurrent-event proportional hazards model in start–stop form, with the baseline stratified by event number and including a subject-specific frailty term to handle within-individual correlation among repeated tests; in the joint model, the same correlation is handled through the shared random effects.</p>
<p>All analyses were performed using R version 4.4.3, with the joint model fitted using the VAJointSurv package, and the standalone Cox and Royston-Parmar flexible parametric models fitted using the survival and rstpm2 packages, respectively.</p>
</sec>
</sec>
<sec id="sec005" sec-type="results">
<title>Results</title>
<p>Among the 506,761 men in the cohort, the median follow-up was 9.8 years, with a total of 1,836,900 PSA tests recorded before censoring at diagnosis (335,172 additional tests were taken after a prostate cancer diagnosis and were excluded). The median number of tests per man was two, and 303,752 men had at least two tests. After censoring at diagnosis, the median age at the first PSA test was 55.8 years. PSA values ranged from near zero to 22,510 ng/mL, with a median value of 1.2 ng/mL (IQR, 0.7–2.6 ng/mL). Overall, 82% of men with repeated PSA testing never reached the age-specific Swedish referral thresholds (≥3 ng/mL for men younger than 70 years, ≥ 5 ng/mL for those aged 70–80 years, and ≥7 ng/mL for those older than 80 years). During follow-up, 30,199 men (6.0%) were diagnosed with prostate cancer; the median age at diagnosis was 69.3 years (IQR, 63.1–75.7), and the median time from the first PSA test to diagnosis was 4.5 years (IQR, 0.0–9.5).</p>
<p><xref ref-type="table" rid="pone.0343751.t001">Table 1</xref> provides an overview of the PSA testing patterns. Total PSA values together with rates for PSA testing, biopsies and diagnosis are higher at increasing ages.</p>
<table-wrap id="pone.0343751.t001" position="float"><object-id pub-id-type="doi">10.1371/journal.pone.0343751.t001</object-id><label>Table 1</label><caption><title>Summary of PSA testing, biopsy, and prostate cancer diagnosis by age group and calendar period, Stockholm 2003-2021.</title></caption>
<alternatives><graphic id="pone.0343751.t001g" mimetype="image" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.t001" xlink:type="simple"/><table><colgroup>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
</colgroup>
<thead>
<tr>
<th align="left">Period</th>
<th align="left">Age group (years)</th>
<th align="left">Person-years (PY) (x1,000)</th>
<th align="left">Tests total (x1,000)</th>
<th align="left">Tests per PY</th>
<th align="left">PSA, Median (IQR), ng/mL</th>
<th align="left">Biopsies total</th>
<th align="left">Biopsies per PY</th>
<th align="left">Diagnosed total</th>
<th align="left">Diagnosed per PY</th>
</tr>
</thead>
<tbody>
<tr>
<td align="left" rowspan="5">2003–2008</td>
<td align="left">&lt;50</td>
<td align="left">268.39</td>
<td align="left">96.96</td>
<td align="left">0.36</td>
<td align="left">0.70 (0.50-1.00)</td>
<td align="left">1,802</td>
<td align="left">0.007</td>
<td align="left">1,254</td>
<td align="left">0.005</td>
</tr>
<tr>
<td align="left">50–59</td>
<td align="left">269.34</td>
<td align="left">174.01</td>
<td align="left">0.65</td>
<td align="left">0.90 (0.60-1.60)</td>
<td align="left">7,966</td>
<td align="left">0.030</td>
<td align="left">6,743</td>
<td align="left">0.025</td>
</tr>
<tr>
<td align="left">60–69</td>
<td align="left">265.72</td>
<td align="left">190.76</td>
<td align="left">0.72</td>
<td align="left">1.40 (0.80-2.80)</td>
<td align="left">10,726</td>
<td align="left">0.040</td>
<td align="left">8,744</td>
<td align="left">0.033</td>
</tr>
<tr>
<td align="left">70–79</td>
<td align="left">152.73</td>
<td align="left">113.46</td>
<td align="left">0.74</td>
<td align="left">2.50 (1.20-5.00)</td>
<td align="left">6,109</td>
<td align="left">0.040</td>
<td align="left">4,646</td>
<td align="left">0.030</td>
</tr>
<tr>
<td align="left">80+</td>
<td align="left">88.39</td>
<td align="left">61.51</td>
<td align="left">0.70</td>
<td align="left">3.40 (1.40-7.90)</td>
<td align="left">2,086</td>
<td align="left">0.024</td>
<td align="left">1,694</td>
<td align="left">0.019</td>
</tr>
<tr>
<td align="left" rowspan="5">2009–2014</td>
<td align="left">&lt;50</td>
<td align="left">585.01</td>
<td align="left">121.05</td>
<td align="left">0.21</td>
<td align="left">0.73 (0.50-1.10)</td>
<td align="left">1,176</td>
<td align="left">0.002</td>
<td align="left">635</td>
<td align="left">0.001</td>
</tr>
<tr>
<td align="left">50–59</td>
<td align="left">733.51</td>
<td align="left">215.78</td>
<td align="left">0.29</td>
<td align="left">0.97 (0.60-1.60)</td>
<td align="left">6,696</td>
<td align="left">0.009</td>
<td align="left">4,094</td>
<td align="left">0.006</td>
</tr>
<tr>
<td align="left">60–69</td>
<td align="left">625.08</td>
<td align="left">259.24</td>
<td align="left">0.41</td>
<td align="left">1.50 (0.81-2.80)</td>
<td align="left">13,740</td>
<td align="left">0.022</td>
<td align="left">7,858</td>
<td align="left">0.013</td>
</tr>
<tr>
<td align="left">70–79</td>
<td align="left">274.34</td>
<td align="left">140.31</td>
<td align="left">0.51</td>
<td align="left">2.20 (1.10-4.30)</td>
<td align="left">7,950</td>
<td align="left">0.029</td>
<td align="left">4,163</td>
<td align="left">0.015</td>
</tr>
<tr>
<td align="left">80+</td>
<td align="left">133.12</td>
<td align="left">66.33</td>
<td align="left">0.50</td>
<td align="left">3.30 (1.43-7.20)</td>
<td align="left">2,752</td>
<td align="left">0.021</td>
<td align="left">1,490</td>
<td align="left">0.011</td>
</tr>
<tr>
<td align="left" rowspan="5">2015–2021</td>
<td align="left">&lt;50</td>
<td align="left">840.84</td>
<td align="left">157.57</td>
<td align="left">0.19</td>
<td align="left">0.69 (0.47-1.00)</td>
<td align="left">504</td>
<td align="left">0.001</td>
<td align="left">222</td>
<td align="left">0.000</td>
</tr>
<tr>
<td align="left">50–59</td>
<td align="left">799.40</td>
<td align="left">234.64</td>
<td align="left">0.29</td>
<td align="left">0.94 (0.58-1.60)</td>
<td align="left">3,551</td>
<td align="left">0.004</td>
<td align="left">1,529</td>
<td align="left">0.002</td>
</tr>
<tr>
<td align="left">60–69</td>
<td align="left">553.09</td>
<td align="left">241.05</td>
<td align="left">0.44</td>
<td align="left">1.50 (0.80-2.80)</td>
<td align="left">8,509</td>
<td align="left">0.015</td>
<td align="left">3,238</td>
<td align="left">0.006</td>
</tr>
<tr>
<td align="left">70–79</td>
<td align="left">405.96</td>
<td align="left">198.72</td>
<td align="left">0.49</td>
<td align="left">2.00 (1.00-3.70)</td>
<td align="left">9,010</td>
<td align="left">0.022</td>
<td align="left">3,158</td>
<td align="left">0.008</td>
</tr>
<tr>
<td align="left">80+</td>
<td align="left">154.61</td>
<td align="left">72.28</td>
<td align="left">0.47</td>
<td align="left">2.70 (1.20-6.00)</td>
<td align="left">3,198</td>
<td align="left">0.021</td>
<td align="left">1,046</td>
<td align="left">0.007</td>
</tr>
</tbody>
</table>
</alternatives></table-wrap>
<sec id="sec006">
<title>Joint model</title>
<p><xref ref-type="fig" rid="pone.0343751.g001">Fig 1</xref> displays the population-level estimates with age as the time scale. Looking at the three panels, we can see the three processes that form the core of the joint model: the longitudinal evolution of PSA (left), the age-specific hazard of undergoing a PSA test (middle), and the age-specific hazard of prostate cancer diagnosis (right). The green curves show population-level estimates with 95% pointwise prediction intervals that account for both fixed-effect uncertainty and between-individual variability.</p>
<fig id="pone.0343751.g001" position="float"><object-id pub-id-type="doi">10.1371/journal.pone.0343751.g001</object-id><label>Fig 1</label><caption><title>Estimated population-level longitudinal and time-to-event processes from the joint model.</title></caption>
<graphic mimetype="image" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.g001" xlink:type="simple"/></fig>
</sec>
<sec id="sec007">
<title>Longitudinal submodel</title>
<p>The longitudinal submodel showed that population-average log PSA values increased nonlinearly with age. There was substantial heterogeneity in PSA trajectories across men, reflected by the large random-effect variances. The standard deviation of the random slope was 677.95 (95% CI: 674.74–681.17; P &lt; .001), and the standard deviation of the random intercept was 0.812 (95% CI: 0.810–0.815; P &lt; .001). The correlation between the intercept and slope was high at 0.59 (95% CI: 0.59–0.60; P &lt; .001), indicating that men with higher baseline PSA tended to have steeper age-related increases. Additionally, we observed a steady increase in the standard deviation of the residuals (<xref ref-type="fig" rid="pone.0343751.g002">Fig 2</xref>), indicating heteroscedasticity. In other words, the magnitude of unexplained variation in log PSA grows larger with age.</p>
<fig id="pone.0343751.g002" position="float"><object-id pub-id-type="doi">10.1371/journal.pone.0343751.g002</object-id><label>Fig 2</label><caption><title>Heatmap (left) and standard deviations (right) of residuals from the longitudinal component of the joint model for log-transformed PSA by age at test.</title></caption>
<graphic mimetype="image" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.g002" xlink:type="simple"/></fig>
<p>These findings were consistent with the standalone longitudinal model (<xref ref-type="supplementary-material" rid="pone.0343751.s001">S1 File</xref>), although the joint model estimated a slightly lower population-average trajectory.</p>
</sec>
<sec id="sec008">
<title>Time-to-event submodels</title>
<p>A doubling of PSA concentration was associated with a higher hazard of diagnosis. This estimated effect was stronger than that obtained when modelling time to diagnosis separately; the Cox proportional hazards model specified using counting-processes formulation with log PSA as a time-updated covariate and robust standard errors clustered by individual, yielded an attenuated estimate. The Royston-Parmar flexible parametric model for time to diagnosis gave a hazard ratio per doubling of PSA of 1.61 (95% CI: 1.59–1.62; P &lt; .001), identical to the Cox estimate.</p>
<p>For the observational process, the joint model indicated that doubling of PSA was positively associated with the hazard of a subsequent test. This was consistent with estimates from the PWP model fitted in isolation. Joint and standalone estimates for both processes are reported in <xref ref-type="table" rid="pone.0343751.t002">Table 2</xref>.</p>
<table-wrap id="pone.0343751.t002" position="float"><object-id pub-id-type="doi">10.1371/journal.pone.0343751.t002</object-id><label>Table 2</label><caption><title>Hazard ratio per doubling of PSA from joint and standalone models for the diagnosis and retesting processes.</title></caption>
<alternatives><graphic id="pone.0343751.t002g" mimetype="image" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.t002" xlink:type="simple"/><table><colgroup>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
<col align="left" valign="middle"/>
</colgroup>
<thead>
<tr>
<th align="left" rowspan="2">Process</th>
<th align="left" colspan="2">Standalone model</th>
<th align="left" colspan="2">Joint model</th>
</tr>
<tr>
<th align="left">HR (95% CI)</th>
<th align="left">P-value</th>
<th align="left">HR (95% CI)</th>
<th align="left">P-value</th>
</tr>
</thead>
<tbody>
<tr>
<td align="left">Time-to-diagnosis</td>
<td align="left">1.61 (1.59-1.62)</td>
<td align="left">&lt;0.001</td>
<td align="left">2.01 (1.99-2.02)</td>
<td align="left">&lt;0.001</td>
</tr>
<tr>
<td align="left">Time-to-test</td>
<td align="left">1.068 (1.066-1.07)</td>
<td align="left">&lt;0.001</td>
<td align="left">1.163 (1.161-1.165)</td>
<td align="left">&lt;0.001</td>
</tr>
</tbody>
</table>
</alternatives></table-wrap>
</sec>
<sec id="sec009">
<title>Patient-specific estimated trajectories</title>
<p><xref ref-type="fig" rid="pone.0343751.g003">Figs 3</xref> and <xref ref-type="fig" rid="pone.0343751.g004">4</xref> illustrate fitted trajectories from the joint model for two individuals who were not diagnosed and diagnosed with prostate cancer, respectively. The green curves are population-level estimates from <xref ref-type="fig" rid="pone.0343751.g001">Fig 1</xref>. The orange curves represent individual-specific estimates based on the conditional modes of the random effects, with orange shaded areas representing 95% confidence intervals. For the diagnosed individual (<xref ref-type="fig" rid="pone.0343751.g004">Fig 4</xref>), the steadily rising PSA trajectory is accompanied by a sharp increase in both the hazard of subsequent testing and the hazard of diagnosis. In contrast, the undiagnosed individual (<xref ref-type="fig" rid="pone.0343751.g003">Fig 3</xref>) had PSA values that closely follow the population-average trajectory, with corresponding hazards for testing and diagnosis remaining consistent with population-average values. The individual-specific confidence intervals narrow across all three processes as more PSA measurements accumulate, most noticeably in the intervals where the tests are taken. It is important to note that these (orange) bands are within-sample posterior estimates conditional on the entire set of observed PSA values for that individual; they are not out-of-sample predictions and have not been validated as such (see Discussion for limitations on individual-level use). The individuals shown in <xref ref-type="fig" rid="pone.0343751.g003">Figs 3</xref> and <xref ref-type="fig" rid="pone.0343751.g004">4</xref> are synthetic; the fitted curves were obtained from the joint model on the original data.</p>
<fig id="pone.0343751.g003" position="float"><object-id pub-id-type="doi">10.1371/journal.pone.0343751.g003</object-id><label>Fig 3</label><caption><title>Estimated longitudinal and time-to-event processes from the joint model for an undiagnosed synthetic individual.</title></caption>
<graphic mimetype="image" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.g003" xlink:type="simple"/></fig>
<fig id="pone.0343751.g004" position="float"><object-id pub-id-type="doi">10.1371/journal.pone.0343751.g004</object-id><label>Fig 4</label><caption><title>Estimated longitudinal and time-to-event processes from the joint model for a diagnosed synthetic individual; the vertical line represents the age at diagnosis.</title></caption>
<graphic mimetype="image" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.g004" xlink:type="simple"/></fig>
</sec>
</sec>
<sec id="sec010" sec-type="conclusions">
<title>Discussion</title>
<p>In this large, population-based study of more than half a million men in Stockholm County, we described how PSA values change with age, the observational process of retesting, and the time to prostate cancer diagnosis. Importantly, comparisons with isolated models demonstrated that standard approaches substantially underestimated the association between PSA and diagnosis by failing to account for the interdependence between biomarker trajectory, testing frequency, and cancer risk.</p>
<p>PSA values increased non-linearly with age and were highly heterogeneous across and within men. We observed considerable heteroscedasticity, with the variability in PSA values increasing with age. This pattern partially explains why using a single PSA threshold for screening in large populations is challenging.</p>
<p>Our comparison of the joint model and the linear mixed model for PSA trajectories highlights the importance of accounting for the observation process when analysing clinical biomarkers obtained through opportunistic or patient-driven testing. Men with elevated PSA levels are more likely to be re-tested, leading to an informative observation process in which the timing and frequency of PSA measurements depend on the underlying PSA trajectory. A mixed effects model assumes that measurement times are independent of the outcome, and consequently can lead to biased estimates. By jointly modelling the PSA process and the observation process, the joint model explicitly accounts for this dependency, therefore potentially providing more realistic estimates of population-level PSA trajectories. In our analysis, this adjustment resulted in slightly lower estimated mean PSA levels compared with the linear mixed-effects model, consistent with the expectation that failure to adjust for informative observation tends to overestimate PSA at the population level.</p>
<p>The mean age at diagnosis was 70.8 years, which is consistent with the higher PSA values observed among men aged 60 and older, likely reflecting cancers that had already developed by that age. However, clear differences in PSA distributions between diagnosed and undiagnosed men were also evident at younger ages.</p>
<p>When the longitudinal PSA process and the event processes were modelled separately, the estimated hazards for diagnosis and retesting were attenuated. For example, the hazard ratio for prostate cancer diagnosis per doubling of PSA was 1.61 in the separate Cox model but increased to 2.01 in the joint model. Similarly, the association between PSA and the risk of retesting rose from HR = 1.068 in the separate PWP model to HR = 1.163 under joint modelling. This attenuation in the separate analyses likely reflects informative observation bias: the timing of PSA measurements is not independent of the biomarker trajectory or the disease progression. When such feedback is ignored, part of the association between PSA and the outcome is absorbed into the unmodelled observation mechanism, leading to an underestimation of effect sizes. The joint model corrects for this by sharing random effects across the longitudinal and survival submodels, therefore accounting for the correlation between PSA levels, testing frequency, and diagnosis risk.</p>
<p>We used variational approximations to fit the joint model. This approach scaled well to a moderately large dataset and produced patient-level posterior estimates of the random effects as a by-product of fitting. These posterior estimates summarise how an individual deviates from the population-average PSA trajectory and come with measures of uncertainty derived from the variational parameters. They are, however, within-sample, conditional on each man’s full set of observed PSA values, and have not been developed or evaluated as a prediction tool in this study. They could form the basis of an individual-level prediction model in future work; such a model would additionally require a held-out validation sample not used in model fitting and an evaluation of calibration, discrimination, and clinical utility, none of which we report here. The patient-specific curves in <xref ref-type="fig" rid="pone.0343751.g003">Figs 3</xref> and <xref ref-type="fig" rid="pone.0343751.g004">4</xref> should therefore be read as illustrations of the fitted random-effects structure rather than as validated predictions.</p>
<p>This study has several limitations. First, we do not explicitly model for a change of PSA trajectory after prostate cancer onset. Although cancer onset is not observable, inspection of individual PSA trajectories suggested that men who were subsequently diagnosed often experienced a more rapid rise in PSA after a certain point. To partially address this issue, we have censored from a prostate cancer diagnosis. Second, the data are observational, leading to substantial variation in the observation process across individuals. Our joint modelling approach seeks to explicitly model for this variation. Third, we lack information on other factors that can elevate PSA, such as urinary tract infections or lower urinary tract symptoms; the absence of these data may contribute to unexplained variability in PSA trajectories. Fourth, our findings may be specific to the Stockholm population. We posit that the connections between the three processes will be similar across populations. It would be useful to perform similar modelling in another population to test this hypothesis. A further limitation is that calendar time was not incorporated into the model. The estimates we report should therefore be interpreted as averaged over the 2003–2020 study period. As a sensitivity check, we refitted standalone time-to-event models within each of the three calendar periods used in <xref ref-type="table" rid="pone.0343751.t001">Table 1</xref>. The hazard ratio per doubling of PSA was in the same direction and of similar magnitude in every period: for diagnosis, 1.71 (95% CI: 1.69–1.74; P &lt; .001), 1.60 (95% CI: 1.58–1.62; P &lt; .001), and 1.57 (95% CI: 1.54–1.59; P &lt; .001); for retesting, 1.09 (95% CI: 1.09–1.09; P &lt; .001), 1.08 (95% CI: 1.08–1.08; P &lt; .001), and 1.05 (95% CI: 1.05–1.05; P &lt; .001), in 2003–2008, 2009–2014, and 2015–2021, respectively. The modest attenuation across periods is consistent with changes in clinical practice during the study window, including the introduction of pre-biopsy MRI and changes in opportunistic testing intensity, but a clear PSA-event association is present in every period. Full details are given in <xref ref-type="supplementary-material" rid="pone.0343751.s001">S1 File</xref>.</p>
<p>Cancer onset is not observed in the data, but it has consequences for the PSA trajectory. When a cancer is present in the prostate, it leaks more PSA into the blood, so PSA levels in a man with undiagnosed cancer tend to rise faster than they would in a cancer-free man of the same age. The current joint model does not account for cancer onset. As a result, men who develop cancer during follow-up but have not yet been diagnosed contribute to the same average PSA trajectory as cancer-free men. Their faster PSA rise is absorbed by their individual random effects rather than being attributed to an underlying biological process.</p>
<p>Gulati et al. [<xref ref-type="bibr" rid="pone.0343751.ref028">28</xref>] addressed this by modelling log PSA as a piecewise linear function of age, with a single slope before cancer onset and a steeper slope after onset. Adapting this idea to our joint model would let us separate two effects that are currently mixed together: how PSA changes with age in cancer-free men, and how it changes after cancer develops. This extension is challenging because cancer onset times are not directly observed and enter the model as additional random effects to be inferred.</p>
<p>We did not account for clustering by general practitioner, clinic, or municipality. A multilevel analysis of PSA testing by municipality or health care provider would be interesting extensions.</p>
<p>In future work, we anticipate that this model can be used to estimate a detailed profile for PSA dynamics for use in a prostate cancer screening model, and estimate PSA uptake and rescreening to describe current opportunistic testing in a population. Additionally, we plan to first estimate the cancer onset distribution using a separate screening model, and then incorporate this estimated distribution into a changepoint model for the longitudinal PSA process.</p>
</sec>
<sec id="sec011" sec-type="conclusions">
<title>Conclusions</title>
<p>PSA trajectories, retesting behaviour, and prostate cancer diagnosis are strongly interdependent, and modelling them separately can substantially underestimate associations with PSA. Joint modelling accounts for informative observation in opportunistic testing and yields more accurate population-level estimates of all three processes. These findings support integrating PSA history and testing intensity into screening models to improve risk stratification and precision.</p>
</sec>
<sec id="sec012" sec-type="supplementary-material">
<title>Supporting information</title>
<supplementary-material id="pone.0343751.s001" mimetype="application/vnd.openxmlformats-officedocument.wordprocessingml.document" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.s001" xlink:type="simple">
<label>S1 File</label>
<caption>
<title>Supplementary methods, figures, and tables.</title>
<p>This file presents seasonal patterns in PSA testing volume (Fig S1); the longitudinal PSA process, including PSA probability densities by age group and by diagnosis status and a comparison of fixed-effect PSA trajectories from the joint and linear mixed models (Figs S2–S4); the observation (retesting) process (Fig S5); disease progression, including Kaplan–Meier estimates, Schoenfeld residuals for log PSA, and age-specific hazard ratios from a piecewise Cox model (Figs S6–S8 and Table S1); and a calendar-period sensitivity analysis of the standalone time-to-event models (Table S2).</p>
<p>(DOCX)</p>
</caption>
</supplementary-material>
</sec>
</body>
<back>
<ref-list>
<title>References</title>
<ref id="pone.0343751.ref001"><label>1</label><mixed-citation publication-type="book" xlink:type="simple"><collab>European Union</collab>. <source>ECIS - European Cancer Information System. Incidence and Mortality 2020</source>. <year>2023</year>.</mixed-citation></ref>
<ref id="pone.0343751.ref002"><label>2</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Ilic</surname> <given-names>D</given-names></name>, <name name-style="western"><surname>Djulbegovic</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Jung</surname> <given-names>JH</given-names></name>, <name name-style="western"><surname>Hwang</surname> <given-names>EC</given-names></name>, <name name-style="western"><surname>Zhou</surname> <given-names>Q</given-names></name>, <name name-style="western"><surname>Cleves</surname> <given-names>A</given-names></name>, <etal>et al</etal>. <article-title>Prostate cancer screening with prostate-specific antigen (PSA) test: a systematic review and meta-analysis</article-title>. <source>BMJ</source>. <year>2018</year>;:k3519. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1136/bmj.k3519" xlink:type="simple">10.1136/bmj.k3519</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref003"><label>3</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Lin</surname> <given-names>K</given-names></name>, <name name-style="western"><surname>Lipsitz</surname> <given-names>R</given-names></name>, <name name-style="western"><surname>Miller</surname> <given-names>T</given-names></name>, <name name-style="western"><surname>Janakiraman</surname> <given-names>S</given-names></name>, <name name-style="western"><surname>U.S. Preventive Services Task</surname> <given-names>Force</given-names></name>. <article-title>Benefits and harms of prostate-specific antigen screening for prostate cancer: an evidence update for the U.S. Preventive Services Task Force</article-title>. <source>Ann Intern Med</source>. <year>2008</year>;<volume>149</volume>(<issue>3</issue>):<fpage>192</fpage>–<lpage>9</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7326/0003-4819-149-3-200808050-00009" xlink:type="simple">10.7326/0003-4819-149-3-200808050-00009</ext-link></comment> <object-id pub-id-type="pmid">18678846</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref004"><label>4</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Vickers</surname> <given-names>AJ</given-names></name>. <article-title>Prostate Cancer Screening: Time to Question How to Optimize the Ratio of Benefits and Harms</article-title>. <source>Ann Intern Med</source>. <year>2017</year>;<volume>167</volume>(<issue>7</issue>):<fpage>509</fpage>–<lpage>10</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7326/M17-2012" xlink:type="simple">10.7326/M17-2012</ext-link></comment> <object-id pub-id-type="pmid">28869975</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref005"><label>5</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Roobol</surname> <given-names>MJ</given-names></name>, <name name-style="western"><surname>de Vos</surname> <given-names>II</given-names></name>, <name name-style="western"><surname>Månsson</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Godtman</surname> <given-names>RA</given-names></name>, <name name-style="western"><surname>Talala</surname> <given-names>KM</given-names></name>, <name name-style="western"><surname>den Hond</surname> <given-names>E</given-names></name>, <etal>et al</etal>. <article-title>European Study of Prostate Cancer Screening — 23-Year Follow-Up</article-title>. <source>New England Journal of Medicine</source>. <year>2025</year>;<volume>393</volume>:<fpage>1669</fpage>–<lpage>80</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1056/NEJMoa2503223" xlink:type="simple">10.1056/NEJMoa2503223</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref006"><label>6</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Martin</surname> <given-names>RM</given-names></name>, <name name-style="western"><surname>Turner</surname> <given-names>EL</given-names></name>, <name name-style="western"><surname>Young</surname> <given-names>GJ</given-names></name>, <name name-style="western"><surname>Metcalfe</surname> <given-names>C</given-names></name>, <name name-style="western"><surname>Walsh</surname> <given-names>EI</given-names></name>, <name name-style="western"><surname>Lane</surname> <given-names>JA</given-names></name>, <etal>et al</etal>. <article-title>Prostate-Specific Antigen Screening and 15-Year Prostate Cancer Mortality: A Secondary Analysis of the CAP Randomized Clinical Trial</article-title>. <source>JAMA</source>. <year>2024</year>;<volume>331</volume>(<issue>17</issue>):<fpage>1460</fpage>–<lpage>70</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1001/jama.2024.4011" xlink:type="simple">10.1001/jama.2024.4011</ext-link></comment> <object-id pub-id-type="pmid">38581198</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref007"><label>7</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Pinsky</surname> <given-names>PF</given-names></name>, <name name-style="western"><surname>Prorok</surname> <given-names>PC</given-names></name>, <name name-style="western"><surname>Yu</surname> <given-names>K</given-names></name>, <name name-style="western"><surname>Kramer</surname> <given-names>BS</given-names></name>, <name name-style="western"><surname>Black</surname> <given-names>A</given-names></name>, <name name-style="western"><surname>Gohagan</surname> <given-names>JK</given-names></name>, <etal>et al</etal>. <article-title>Extended mortality results for prostate cancer screening in the PLCO trial with median follow-up of 15 years</article-title>. <source>Cancer</source>. <year>2017</year>;<volume>123</volume>(<issue>4</issue>):<fpage>592</fpage>–<lpage>9</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1002/cncr.30474" xlink:type="simple">10.1002/cncr.30474</ext-link></comment> <object-id pub-id-type="pmid">27911486</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref008"><label>8</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Catalona</surname> <given-names>WJ</given-names></name>, <name name-style="western"><surname>Southwick</surname> <given-names>PC</given-names></name>, <name name-style="western"><surname>Slawin</surname> <given-names>KM</given-names></name>, <name name-style="western"><surname>Partin</surname> <given-names>AW</given-names></name>, <name name-style="western"><surname>Brawer</surname> <given-names>MK</given-names></name>, <name name-style="western"><surname>Flanigan</surname> <given-names>RC</given-names></name>, <etal>et al</etal>. <article-title>Comparison of percent free PSA, PSA density, and age-specific PSA cutoffs for prostate cancer detection and staging</article-title>. <source>Urology</source>. <year>2000</year>;<volume>56</volume>(<issue>2</issue>):<fpage>255</fpage>–<lpage>60</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/s0090-4295(00)00637-3" xlink:type="simple">10.1016/s0090-4295(00)00637-3</ext-link></comment> <object-id pub-id-type="pmid">10925089</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref009"><label>9</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Luboldt</surname> <given-names>H-J</given-names></name>, <name name-style="western"><surname>Schindler</surname> <given-names>JF</given-names></name>, <name name-style="western"><surname>Rübben</surname> <given-names>H</given-names></name>. <article-title>Age-Specific Reference Ranges for Prostate-Specific Antigen as a Marker for Prostate Cancer</article-title>. <source>EAU-EBU Update Series</source>. <year>2007</year>;<volume>5</volume>(<issue>1</issue>):<fpage>38</fpage>–<lpage>48</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.eeus.2006.10.003" xlink:type="simple">10.1016/j.eeus.2006.10.003</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref010"><label>10</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Moul</surname> <given-names>JW</given-names></name>, <name name-style="western"><surname>Sun</surname> <given-names>L</given-names></name>, <name name-style="western"><surname>Hotaling</surname> <given-names>JM</given-names></name>, <name name-style="western"><surname>Fitzsimons</surname> <given-names>NJ</given-names></name>, <name name-style="western"><surname>Polascik</surname> <given-names>TJ</given-names></name>, <name name-style="western"><surname>Robertson</surname> <given-names>CN</given-names></name>, <etal>et al</etal>. <article-title>Age adjusted prostate specific antigen and prostate specific antigen velocity cut points in prostate cancer screening</article-title>. <source>J Urol</source>. <year>2007</year>;<volume>177</volume>(<issue>2</issue>):<fpage>499</fpage>–<lpage>503</lpage>; discussion 503-4. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.juro.2006.09.063" xlink:type="simple">10.1016/j.juro.2006.09.063</ext-link></comment> <object-id pub-id-type="pmid">17222618</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref011"><label>11</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Carter</surname> <given-names>HB</given-names></name>, <name name-style="western"><surname>Pearson</surname> <given-names>JD</given-names></name>, <name name-style="western"><surname>Metter</surname> <given-names>EJ</given-names></name>, <name name-style="western"><surname>Brant</surname> <given-names>LJ</given-names></name>, <name name-style="western"><surname>Chan</surname> <given-names>DW</given-names></name>, <name name-style="western"><surname>Andres</surname> <given-names>R</given-names></name>, <etal>et al</etal>. <article-title>Longitudinal evaluation of prostate-specific antigen levels in men with and without prostate disease</article-title>. <source>JAMA</source>. <year>1992</year>;<volume>267</volume>(<issue>16</issue>):<fpage>2215</fpage>–<lpage>20</lpage>. <object-id pub-id-type="pmid">1372942</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref012"><label>12</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Catalona</surname> <given-names>WJ</given-names></name>, <name name-style="western"><surname>Smith</surname> <given-names>DS</given-names></name>, <name name-style="western"><surname>Ratliff</surname> <given-names>TL</given-names></name>, <name name-style="western"><surname>Dodds</surname> <given-names>KM</given-names></name>, <name name-style="western"><surname>Coplen</surname> <given-names>DE</given-names></name>, <name name-style="western"><surname>Yuan</surname> <given-names>JJ</given-names></name>, <etal>et al</etal>. <article-title>Measurement of prostate-specific antigen in serum as a screening test for prostate cancer</article-title>. <source>N Engl J Med</source>. <year>1991</year>;<volume>324</volume>(<issue>17</issue>):<fpage>1156</fpage>–<lpage>61</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1056/NEJM199104253241702" xlink:type="simple">10.1056/NEJM199104253241702</ext-link></comment> <object-id pub-id-type="pmid">1707140</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref013"><label>13</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Vickers</surname> <given-names>AJ</given-names></name>, <name name-style="western"><surname>Thompson</surname> <given-names>IM</given-names></name>, <name name-style="western"><surname>Klein</surname> <given-names>E</given-names></name>, <name name-style="western"><surname>Carroll</surname> <given-names>PR</given-names></name>, <name name-style="western"><surname>Scardino</surname> <given-names>PT</given-names></name>. <article-title>A commentary on PSA velocity and doubling time for clinical decisions in prostate cancer</article-title>. <source>Urology</source>. <year>2014</year>;<volume>83</volume>(<issue>3</issue>):<fpage>592</fpage>–<lpage>6</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.urology.2013.09.075" xlink:type="simple">10.1016/j.urology.2013.09.075</ext-link></comment> <object-id pub-id-type="pmid">24581521</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref014"><label>14</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Vickers</surname> <given-names>AJ</given-names></name>, <name name-style="western"><surname>Brewster</surname> <given-names>SF</given-names></name>. <article-title>PSA Velocity and Doubling Time in Diagnosis and Prognosis of Prostate Cancer</article-title>. <source>Br J Med Surg Urol</source>. <year>2012</year>;<volume>5</volume>(<issue>4</issue>):<fpage>162</fpage>–<lpage>8</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.bjmsu.2011.08.006" xlink:type="simple">10.1016/j.bjmsu.2011.08.006</ext-link></comment> <object-id pub-id-type="pmid">22712027</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref015"><label>15</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Wallner</surname> <given-names>LP</given-names></name>, <name name-style="western"><surname>Frencher</surname> <given-names>SK</given-names></name>, <name name-style="western"><surname>Hsu</surname> <given-names>J-WY</given-names></name>, <name name-style="western"><surname>Chao</surname> <given-names>CR</given-names></name>, <name name-style="western"><surname>Nichol</surname> <given-names>MB</given-names></name>, <name name-style="western"><surname>Loo</surname> <given-names>RK</given-names></name>, <etal>et al</etal>. <article-title>Changes in serum prostate-specific antigen levels and the identification of prostate cancer in a large managed care population</article-title>. <source>BJU Int</source>. <year>2013</year>;<volume>111</volume>(<issue>8</issue>):<fpage>1245</fpage>–<lpage>52</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1111/j.1464-410X.2012.11651.x" xlink:type="simple">10.1111/j.1464-410X.2012.11651.x</ext-link></comment> <object-id pub-id-type="pmid">23320750</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref016"><label>16</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Karnes</surname> <given-names>RJ</given-names></name>, <name name-style="western"><surname>MacKintosh</surname> <given-names>FR</given-names></name>, <name name-style="western"><surname>Morrell</surname> <given-names>CH</given-names></name>, <name name-style="western"><surname>Rawson</surname> <given-names>L</given-names></name>, <name name-style="western"><surname>Sprenkle</surname> <given-names>PC</given-names></name>, <name name-style="western"><surname>Kattan</surname> <given-names>MW</given-names></name>, <etal>et al</etal>. <article-title>Prostate-Specific Antigen Trends Predict the Probability of Prostate Cancer in a Very Large U.S. Veterans Affairs Cohort</article-title>. <source>Front Oncol</source>. <year>2018</year>;<volume>8</volume>:<fpage>296</fpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.3389/fonc.2018.00296" xlink:type="simple">10.3389/fonc.2018.00296</ext-link></comment> <object-id pub-id-type="pmid">30128303</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref017"><label>17</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Kohar</surname> <given-names>A</given-names></name>, <name name-style="western"><surname>Cramb</surname> <given-names>SM</given-names></name>, <name name-style="western"><surname>Pickles</surname> <given-names>K</given-names></name>, <name name-style="western"><surname>Smith</surname> <given-names>DP</given-names></name>, <name name-style="western"><surname>Baade</surname> <given-names>PD</given-names></name>. <article-title>Changes in prostate specific antigen (PSA) “screening” patterns by geographic region and socio-economic status in Australia: Analysis of medicare data in 50-69 year old men</article-title>. <source>Cancer Epidemiol</source>. <year>2023</year>;<volume>83</volume>:<fpage>102338</fpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.canep.2023.102338" xlink:type="simple">10.1016/j.canep.2023.102338</ext-link></comment> <object-id pub-id-type="pmid">36841020</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref018"><label>18</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Nordström</surname> <given-names>T</given-names></name>, <name name-style="western"><surname>Aly</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Clements</surname> <given-names>MS</given-names></name>, <name name-style="western"><surname>Weibull</surname> <given-names>CE</given-names></name>, <name name-style="western"><surname>Adolfsson</surname> <given-names>J</given-names></name>, <name name-style="western"><surname>Grönberg</surname> <given-names>H</given-names></name>. <article-title>Prostate-specific Antigen (PSA) Testing Is Prevalent and Increasing in Stockholm County, Sweden, Despite No Recommendations for PSA Screening: Results from a Population-based Study, 2003–2011</article-title>. <source>Eur Urol</source>. <year>2013</year>;<volume>63</volume>:<fpage>419</fpage>–<lpage>25</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/J.EURURO.2012.10.001" xlink:type="simple">10.1016/J.EURURO.2012.10.001</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref019"><label>19</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Mariotto</surname> <given-names>AB</given-names></name>, <name name-style="western"><surname>Etzioni</surname> <given-names>R</given-names></name>, <name name-style="western"><surname>Krapcho</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Feuer</surname> <given-names>EJ</given-names></name>. <article-title>Reconstructing PSA testing patterns between black and white men in the US from Medicare claims and the National Health Interview Survey</article-title>. <source>Cancer</source>. <year>2007</year>;<volume>109</volume>(<issue>9</issue>):<fpage>1877</fpage>–<lpage>86</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1002/cncr.22607" xlink:type="simple">10.1002/cncr.22607</ext-link></comment> <object-id pub-id-type="pmid">17372918</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref020"><label>20</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Collins</surname> <given-names>KK</given-names></name>, <name name-style="western"><surname>Oke</surname> <given-names>JL</given-names></name>, <name name-style="western"><surname>Virdee</surname> <given-names>PS</given-names></name>, <name name-style="western"><surname>Perera</surname> <given-names>R</given-names></name>, <name name-style="western"><surname>Nicholson</surname> <given-names>BD</given-names></name>. <article-title>Prostate specific antigen retesting intervals and trends in England: population based cohort study</article-title>. <source>BMJ</source>. <year>2025</year>;<volume>391</volume>:e083800. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1136/bmj-2024-083800" xlink:type="simple">10.1136/bmj-2024-083800</ext-link></comment> <object-id pub-id-type="pmid">41062172</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref021"><label>21</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Ferrer</surname> <given-names>L</given-names></name>, <name name-style="western"><surname>Rondeau</surname> <given-names>V</given-names></name>, <name name-style="western"><surname>Dignam</surname> <given-names>J</given-names></name>, <name name-style="western"><surname>Pickles</surname> <given-names>T</given-names></name>, <name name-style="western"><surname>Jacqmin-Gadda</surname> <given-names>H</given-names></name>, <name name-style="western"><surname>Proust-Lima</surname> <given-names>C</given-names></name>. <article-title>Joint modelling of longitudinal and multi-state processes: application to clinical progressions in prostate cancer</article-title>. <source>Stat Med</source>. <year>2016</year>;<volume>35</volume>(<issue>22</issue>):<fpage>3933</fpage>–<lpage>48</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1002/sim.6972" xlink:type="simple">10.1002/sim.6972</ext-link></comment> <object-id pub-id-type="pmid">27090611</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref022"><label>22</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Proust-Lima</surname> <given-names>C</given-names></name>, <name name-style="western"><surname>Séne</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Taylor</surname> <given-names>JMG</given-names></name>, <name name-style="western"><surname>Jacqmin-Gadda</surname> <given-names>H</given-names></name>. <article-title>Joint latent class models for longitudinal and time-to-event data: a review</article-title>. <source>Stat Methods Med Res</source>. <year>2014</year>;<volume>23</volume>(<issue>1</issue>):<fpage>74</fpage>–<lpage>90</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1177/0962280212445839" xlink:type="simple">10.1177/0962280212445839</ext-link></comment> <object-id pub-id-type="pmid">22517270</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref023"><label>23</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Rizopoulos</surname> <given-names>D</given-names></name>, <name name-style="western"><surname>Taylor</surname> <given-names>JM</given-names></name>, <name name-style="western"><surname>Papageorgiou</surname> <given-names>G</given-names></name>, <name name-style="western"><surname>Morgan</surname> <given-names>TM</given-names></name>. <article-title>Using joint models for longitudinal and time-to-event data to investigate the causal effect of salvage therapy after prostatectomy</article-title>. <source>Stat Methods Med Res</source>. <year>2024</year>;<volume>33</volume>(<issue>5</issue>):<fpage>894</fpage>–<lpage>908</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1177/09622802241239003" xlink:type="simple">10.1177/09622802241239003</ext-link></comment> <object-id pub-id-type="pmid">38502034</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref024"><label>24</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Abbadi</surname> <given-names>A</given-names></name>, <name name-style="western"><surname>Eklund</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Aly</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Clements</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Crippa</surname> <given-names>A</given-names></name>, <name name-style="western"><surname>Discacciati</surname> <given-names>A</given-names></name>, <etal>et al</etal>. <article-title>Data Resource Profile: STHLM0, the Stockholm Prostate Cancer Diagnostics Register</article-title>. <source>Int J Epidemiol</source>. <year>2025</year>;<volume>54</volume>(<issue>3</issue>):dyaf062. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1093/ije/dyaf062" xlink:type="simple">10.1093/ije/dyaf062</ext-link></comment> <object-id pub-id-type="pmid">40461103</object-id></mixed-citation></ref>
<ref id="pone.0343751.ref025"><label>25</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Chandra Engel</surname> <given-names>J</given-names></name>, <name name-style="western"><surname>Rai</surname> <given-names>B</given-names></name>, <name name-style="western"><surname>Eklund</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Jäderling</surname> <given-names>F</given-names></name>, <name name-style="western"><surname>Clements</surname> <given-names>M</given-names></name>, <name name-style="western"><surname>Nordström</surname> <given-names>T</given-names></name>. <article-title>Trends and patterns in prostate cancer diagnostics during the era of MRI implementation – real-world evidence from a population-based study in the Stockholm region, Sweden 2010–2023</article-title>. <source>Eur Urol Open Sci</source>. <year>2026</year>;<volume>87</volume>:<fpage>48</fpage>–<lpage>56</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/J.EUROS.2026.03.015" xlink:type="simple">10.1016/J.EUROS.2026.03.015</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref026"><label>26</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Ormerod</surname> <given-names>JT</given-names></name>, <name name-style="western"><surname>Wand</surname> <given-names>MP</given-names></name>. <article-title>Gaussian variational approximate inference for generalized linear mixed models</article-title>. <source>Journal of Computational and Graphical Statistics</source>. <year>2012</year>;<volume>21</volume>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1198/jcgs.2011.09118" xlink:type="simple">10.1198/jcgs.2011.09118</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref027"><label>27</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Christoffersen</surname> <given-names>B</given-names></name>, <name name-style="western"><surname>Humphreys</surname> <given-names>K</given-names></name>, <name name-style="western"><surname>Gasparini</surname> <given-names>A</given-names></name>, <name name-style="western"><surname>Akynkozhayev</surname> <given-names>B</given-names></name>, <name name-style="western"><surname>Kjellström</surname> <given-names>H</given-names></name>, <name name-style="western"><surname>Clements</surname> <given-names>M</given-names></name>. <article-title>Joint models with multiple markers and multiple time-to-event outcomes using variational approximations</article-title>. <source>arXiv</source>. <year>2025</year>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/2512.13962" xlink:type="simple">2512.13962</ext-link></comment></mixed-citation></ref>
<ref id="pone.0343751.ref028"><label>28</label><mixed-citation publication-type="journal" xlink:type="simple"><name name-style="western"><surname>Gulati</surname> <given-names>R</given-names></name>, <name name-style="western"><surname>Inoue</surname> <given-names>L</given-names></name>, <name name-style="western"><surname>Katcher</surname> <given-names>J</given-names></name>, <name name-style="western"><surname>Hazelton</surname> <given-names>W</given-names></name>, <name name-style="western"><surname>Etzioni</surname> <given-names>R</given-names></name>. <article-title>Calibrating disease progression models using population data: a critical precursor to policy development in cancer control</article-title>. <source>Biostatistics</source>. <year>2010</year>;<volume>11</volume>(<issue>4</issue>):<fpage>707</fpage>–<lpage>19</lpage>. <comment>doi: <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1093/biostatistics/kxq036" xlink:type="simple">10.1093/biostatistics/kxq036</ext-link></comment></mixed-citation></ref>
</ref-list>
</back>
<sub-article article-type="aggregated-review-documents" id="pone.0343751.r001" specific-use="decision-letter">
<front-stub>
<article-id pub-id-type="doi">10.1371/journal.pone.0343751.r001</article-id>
<title-group>
<article-title>Decision Letter 0</article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name name-style="western"><surname>Wroński</surname>
<given-names>Stanisław</given-names>
</name>
<role>Academic Editor</role>
</contrib>
</contrib-group>
<permissions>
<copyright-year>2026</copyright-year>
<copyright-holder>Stanisław Wroński</copyright-holder>
<license xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">
<license-p>This is an open access article distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">Creative Commons Attribution License</ext-link>, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.</license-p></license>
</permissions>
<related-object document-id="10.1371/journal.pone.0343751" document-id-type="doi" document-type="article" id="rel-obj001" link-type="peer-reviewed-article"/>
<custom-meta-group>
<custom-meta>
<meta-name>Submission Version</meta-name>
<meta-value>0</meta-value>
</custom-meta>
</custom-meta-group>
</front-stub>
<body>
<p><named-content content-type="letter-date">14 May 2026</named-content></p>
<p>--&gt;PONE-D-26-05935--&gt;--&gt;Joint modelling of PSA dynamics and prostate cancer risks: A population-based study--&gt;--&gt;PLOS One</p>
<p>Dear Dr. Birzhan Akynkozhayev,</p>
<p>Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.</p>
<p>--&gt;--&gt;</p>
<p>Please submit your revised manuscript by 15th June 2026. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at <email xlink:type="simple">plosone@plos.org</email>. When you're ready to submit your revision, log on to <ext-link ext-link-type="uri" xlink:href="https://www.editorialmanager.com/pone/" xlink:type="simple">https://www.editorialmanager.com/pone/</ext-link> and select the 'Submissions Needing Revision' folder to locate your manuscript file.</p>
<p>Please include the following items when submitting your revised manuscript:--&gt;</p>
<p><list list-type="bullet"><list-item><p>A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.</p>
</list-item>
<list-item>
<p>A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.</p>
</list-item>
<list-item>
<p>An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.</p></list-item></list></p>
<p>If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.</p>
<p>If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: <ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols" xlink:type="simple">https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols</ext-link>. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at <ext-link ext-link-type="uri" xlink:href="https://plos.org/protocols?utm_medium=editorial-email&amp;utm_source=authorletters&amp;utm_campaign=protocols" xlink:type="simple">https://plos.org/protocols?utm_medium=editorial-email&amp;utm_source=authorletters&amp;utm_campaign=protocols</ext-link>.</p>
<p>As the corresponding author, your ORCID iD is verified in the submission system and will appear in the published article. PLOS supports the use of ORCID, and we encourage all coauthors to register for an ORCID iD and use it as well. Please encourage your coauthors to verify their ORCID iD within the submission system before final acceptance, as unverified ORCID iDs will not appear in the published article. <italic>Only</italic>  the individual author can complete the verification step; PLOS staff <italic>cannot</italic>  verify ORCID iDs on behalf of authors.</p>
<p>We look forward to receiving your revised manuscript.</p>
<p>Kind regards,</p>
<p>Stanisław Jacek Wroński, M.D., Ph.D, FEBU</p>
<p>Academic Editor</p>
<p>PLOS One</p>
<p>Journal Requirements:</p>
<p>When submitting your revision, we need you to address these additional requirements.</p>
<p>1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at</p>
<p><ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf" xlink:type="simple">https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf</ext-link> and</p>
<p><ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf" xlink:type="simple">https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf</ext-link></p>
<p>2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.</p>
<p>If you are reporting a retrospective study of medical records or archived samples, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.</p>
<p>3. Thank you for stating the following financial disclosure:</p>
<p>"Holly Edward is a recipient of the Canadian Institutes of Health Research’s Doctoral Research Award (Funding Reference Number: 193451)."</p>
<p>Please state what role the funders took in the study.  If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."</p>
<p>If this statement is not correct you must amend it as needed.</p>
<p>Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.</p>
<p>4. In this instance it seems there may be acceptable restrictions in place that prevent the public sharing of your minimal data. However, in line with our goal of ensuring long-term data availability to all interested researchers, PLOS’ Data Policy states that authors cannot be the sole named individuals responsible for ensuring data access (<ext-link ext-link-type="uri" xlink:href="http://journals.plos.org/plosone/s/data-availability#loc-acceptable-data-sharing-methods" xlink:type="simple">http://journals.plos.org/plosone/s/data-availability#loc-acceptable-data-sharing-methods</ext-link>).</p>
<p>Data requests to a non-author institutional point of contact, such as a data access or ethics committee, helps guarantee long term stability and availability of data. Providing interested researchers with a durable point of contact ensures data will be accessible even if an author changes email addresses, institutions, or becomes unavailable to answer requests.</p>
<p>Before we proceed with your manuscript, please also provide non-author contact information (phone/email/hyperlink) for a data access committee, ethics committee, or other institutional body to which data requests may be sent. If no institutional body is available to respond to requests for your minimal data, please consider if there any institutional representatives who did not collaborate in the study, and are not listed as authors on the manuscript, who would be able to hold the data and respond to external requests for data access? If so, please provide their contact information (i.e., email address). Please also provide details on how you will ensure persistent or long-term data storage and availability.</p>
<p>5. Thank you for stating the following in the Competing Interests section:</p>
<p>"I have read the journal's policy and the authors of this manuscript have the following competing interests: Dr Nordström reported owning shares in A3P Biomedical. No other disclosures were reported. This does not alter our adherence to PLOS ONE policies on sharing data and materials."</p>
<p>Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors <ext-link ext-link-type="uri" xlink:href="http://journals.plos.org/plosone/s/competing-interests" xlink:type="simple">http://journals.plos.org/plosone/s/competing-interests).</ext-link>  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.</p>
<p>Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.</p>
<p>6. We note that you have indicated that there are restrictions to data sharing for this study. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see <ext-link ext-link-type="uri" xlink:href="http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions" xlink:type="simple">http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions</ext-link>.</p>
<p>Before we proceed with your manuscript, please address the following prompts:</p>
<p>a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., a Research Ethics Committee or Institutional Review Board, etc.). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.</p>
<p>b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of recommended repositories, please see</p>
<p><ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/recommended-repositories" xlink:type="simple">https://journals.plos.org/plosone/s/recommended-repositories</ext-link>. You also have the option of uploading the data as Supporting Information files, but we would recommend depositing data directly to a data repository if possible.</p>
<p>We will update your Data Availability statement on your behalf to reflect the information you provide.</p>
<p>7. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.</p>
<p>Reviewers' comments:</p>
<p>Reviewer's Responses to Questions</p>
<p>--&gt;<bold>Comments to the Author</bold></p>
<p>1. Is the manuscript technically sound, and do the data support the conclusions?</p>
<p>The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. --&gt;</p>
<p>Reviewer #1: Yes</p>
<p>Reviewer #2: Partly</p>
<p>**********</p>
<p>--&gt;2. Has the statistical analysis been performed appropriately and rigorously? --&gt;</p>
<p>Reviewer #1: Yes</p>
<p>Reviewer #2: Yes</p>
<p>**********</p>
<p>--&gt;3. Have the authors made all data underlying the findings in their manuscript fully available?</p>
<p>The <ext-link ext-link-type="uri" xlink:href="http://www.plosone.org/static/policies.action#sharing" xlink:type="simple">PLOS Data policy</ext-link> requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--&gt;</p>
<p>Reviewer #1: Yes</p>
<p>Reviewer #2: No</p>
<p>**********</p>
<p>--&gt;4. Is the manuscript presented in an intelligible fashion and written in standard English?</p>
<p>PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.--&gt;</p>
<p>Reviewer #1: Yes</p>
<p>Reviewer #2: Yes</p>
<p>**********</p>
<p>--&gt;5. Review Comments to the Author</p>
<p>Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)--&gt;</p>
<p>Reviewer #1: The authors have analyzed PSA testing dynamics by applying a joint model of PSA values and testing patterns and risk of prostate cancer to observational data. The model identified stronger associations between PSA increase and prostate cancer diagnosis and retesting. They concluded that the joint approach is more appropriate compared to separate models and that PSA dynamics may be clinically informative.</p>
<p>The manuscript is well written and brings an important contribution to the field of modelling of PSA dynamics and screening. I have some (mainly minor) comments that may improve the manuscript.</p>
<p>Title</p>
<p>I suggest that you add the target population to the title – for example:</p>
<p>“Joint modelling of PSA dynamics and prostate cancer risks: A population-based study of men without prior prostate cancer diagnosis”</p>
<p>Abstract</p>
<p>Without having read the rest of the manuscript, it is not entirely clear how the conclusions are supported by the results presented in the abstract alone. How does the joint model correct for bias, and in what sense is it more accurate? Could this be demonstrated with results, for example by explicitly comparing it with the separate models?</p>
<p>Minor: I do not agree that the following is a limitation “the model does not explicitly account for PSA trajectory changes after cancer onset” considering that you aimed to model PSA dynamics in men without prior prostate cancer diagnosis. It could perhaps be mentioned as a delimitation in the discussion.</p>
<p>Introduction</p>
<p>Minor: The last paragraph – I suggest that you add the target population to the aim, for example something like this:</p>
<p>“In this study, we aimed to develop a detailed population-based PSA model for men without a prior prostate cancer diagnosis.”</p>
<p>Materials and methods</p>
<p>The authors state that a PSA is followed by an MRI and then possibly by a biopsy. At the same time data on men that underwent PSA testing was collected between 2003-2020. MRI was only introduced into the diagnostic pathway during the latter part of the study window. I suggest that the authors clarify how the guidelines have changed over the study period and provide details on when MRI was introduced.</p>
<p>Minor: Also, not all men should undergo an MRI before biopsy, so I suggest that you say that the contemporary typical diagnostic pathway is PSA – MRI – biopsy.</p>
<p>Minor: I would say that the paragraphs between lines 97 and 113 are results and should be moved to the results section.</p>
<p>Minor: what was the lower detection limit for PSA from the laboratories and did this vary over calendar time? Was there an upper reported limit for PSA? How did you handle this in the modelling?</p>
<p>Given the emphasis on the novelty of the modelling approach, I would have appreciated a bit more detail about the model in the main text under “Statistical methods”. For instance, how were the spline knots chosen, and how was the log transformed PSA trajectory explicitly incorporated into the subsequent Cox models?</p>
<p>Minor: It would also have been good with a sentence or two about how the model was fitted, and a short description of how you plan to illustrate or present the fitted model(s). This would be valuable especially to those unfamiliar with the joint/latent modelling approach.</p>
<p>Why did the separate models not include splines, or did they? It is not clear based on reading lines 135-139. I would have expected the separate models to be as similar as possible to the joint model in terms of how the covariates were represented.</p>
<p>Minor: what was the covariance matrix for the random effects in the longitudinal submodel?</p>
<p>Minor: supplement – I think there should be an explicit reference to the arXiv article that describes the method with the relevant identifier.</p>
<p>Results</p>
<p>The results are clearly presented. However, I believe it would be valuable to include an explicit comparison between the joint model and the separate models in the manuscript, especially since this is a key finding highlighted in both the abstract and the discussion’s summary of results. Perhaps Figures 1 and 3–4 could be combined to make space for additional details on all the models currently found in the supplement.</p>
<p>Minor: how come there is such a wide interval for PSA for younger ages in Figure 1? Is it for the same reason as for the later figures, or is there a larger heterogeneity in younger men?</p>
<p>Minor: did model performance somehow deteriorate or improve over calendar time, considering the changing diagnostic landscape and in particular the introduction of MRI?</p>
<p>Discussion/conclusions</p>
<p>Minor: consider to give an example of how the model can be used in a personalized decision-making context?</p>
<p>I would have appreciated a more in‑depth discussion of the modelling choices and the plausibility of the underlying assumptions. For example, the authors conclude the discussion by stating that they plan to model cancer onset and integrate it into the joint model. It would be helpful for readers to understand why this addition would be valuable—for example, in what ways might it improve model performance? Could the authors illustrate how the current model may underperform, perhaps by showing that it does not fully capture the heterogeneity and dynamics in the data over time, and how the proposed extension might address this limitation?</p>
<p>Reviewer #2: The endeavour to jointly model longitudinal PSA patterns based on age and retesting is of interest. The data used is appropriate and the methodology applied is overall relevant. The manuscript is well written. I have some comments and questions that may help the authors to improve the manuscript:</p>
<p>1. There is a substantial amount of descriptive data regarding PSA values in the methods section. This is more appropriate to present in the results section.</p>
<p>2. How can calendar time effects during the 17 years study period influence the results? Was calendar time handled in the modelling? If not, why not?</p>
<p>3. Can differences in clinical practice within the source population create correlations between different hospitals/clinics/healthcare providers?</p>
<p>4. How was the situation handled when the interval between the available PSA-values from an individual is very short (weeks). This could likely introduce unstable PSA trajectories.</p>
<p>5. What is the rationale for including the entire age range in the study population? What are the consequences of including the apparently unstable data form the extremes of the age distribution (&lt;40 and &gt;90) in the modelling exercises?</p>
<p>6. The finding that PSA doubling time is associated with an increased likelihood for a new test is not unexpected.</p>
<p>7. The expression in the results section that “uncertainty in the earlier parts of each trajectory (before the first measurement) is informed by future observations, providing a retrospective estimate of the patient’s past” is a bit difficult to understand the relevance of. The authors should justify the relevance of predicting a single individual’s observations, when this individual has been part of the observations used to develop the prediction model. It would likely be more relevant to display the prediction against variability of observations, and/or attempt a validation against data not used in model development, to avoid overly enthusiastic assessments of model performance.</p>
<p>8. The figures 3 and 4 display multiple PSA values at specific ages for single individuals, and for the man diagnosed with prostate cancer, the specific age at diagnosis. This appears as information that risk being able to identify specific individuals. Please verify that such presentation of data is compatible with the ethical approval and protection of personal integrity.</p>
<p>9. The results presentation regarding validation of the model predictions appears limited and not in line with guidelines for reporting of prediction models. Please review relevant guidelines and preferably improve adherence to these.</p>
<p>10. In the discussion it is stated that “The mean age at diagnosis was 70.8 years, which explains the higher PSA densities observed among men aged 60 and older, likely reflecting cancers that had already developed by that age”. There is, however, no mentioning of data on PSA density in the methods and results sections. Please clarify.</p>
<p>11. An important concern with the current manuscript is that the results presentation and discussion allude to use of the models for individual level predictions. The validation of the models does not substantiate such claims. If the models are intended for use on the individual level a much more elaborate validation is expected, with robust support for external validity. An important restriction is how to describe and communicate the uncertainty of model predictions on the individual level. As currently presented, claims for individual level predictions have not been justified, and either the analyses should be substantially expanded or these claims be removed. The statement that “This approach scaled well to a moderately large dataset and supported personalised predictions, including measures of uncertainty” is not agreed.</p>
<p>**********</p>
<p>--&gt;6. PLOS authors have the option to publish the peer review history of their article (<ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/editorial-and-peer-review-process#loc-peer-review-history" xlink:type="simple">what does this mean?</ext-link>). If published, this will include your full peer review and any attached files.</p>
<p>If you choose “no”, your identity will remain anonymous but your review may still be made public.</p>
<p><bold>Do you want your identity to be public for this peer review?</bold>  For information about this choice, including consent withdrawal, please see our <ext-link ext-link-type="uri" xlink:href="https://www.plos.org/privacy-policy" xlink:type="simple">Privacy Policy</ext-link>.--&gt;</p>
<p>Reviewer #1: No</p>
<p>Reviewer #2: No</p>
<p>**********</p>
<p>[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]</p>
<p>To ensure your figures meet our technical requirements, please review our figure guidelines: <ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/figures" xlink:type="simple">https://journals.plos.org/plosone/s/figures</ext-link></p>
<p>You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: <ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation" xlink:type="simple">https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.</ext-link></p>
<p>NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.</p>
</body>
</sub-article>
<sub-article article-type="author-comment" id="pone.0343751.r002">
<front-stub>
<article-id pub-id-type="doi">10.1371/journal.pone.0343751.r002</article-id>
<title-group>
<article-title>Author response to Decision Letter 1</article-title>
</title-group>
<related-object document-id="10.1371/journal.pone.0343751" document-id-type="doi" document-type="peer-reviewed-article" id="rel-obj002" link-type="rebutted-decision-letter" object-id="10.1371/journal.pone.0343751.r001" object-id-type="doi" object-type="decision-letter"/>
<custom-meta-group>
<custom-meta>
<meta-name>Submission Version</meta-name>
<meta-value>1</meta-value>
</custom-meta>
</custom-meta-group>
</front-stub>
<body>
<p><named-content content-type="author-response-date">27 May 2026</named-content></p>
<p>Dear Editor,</p>
<p>Thank you for the opportunity to submit our revised manuscript.</p>
<p>As our detailed point-by-point responses to the reviewers' comments contain mathematical formulae and specific formatting that cannot be rendered correctly within this plain text box, we have provided our complete response in the attached "Response_to_Reviewers.pdf" document.</p>
<p>Please also refer to the attached "Cover_Letter_Revision.docx" for our responses to the journal-level requirements.</p>
<p>Thank you for your time and consideration.</p>
<supplementary-material id="pone.0343751.s003" mimetype="application/pdf" position="float" xlink:href="info:doi/10.1371/journal.pone.0343751.s003" xlink:type="simple">
<label>Attachment</label>
<caption>
<p>Submitted filename: <named-content content-type="submitted-filename">Response_to_Reviewers.pdf</named-content></p>
</caption>
</supplementary-material>
</body>
</sub-article>
<sub-article article-type="aggregated-review-documents" id="pone.0343751.r003" specific-use="decision-letter">
<front-stub>
<article-id pub-id-type="doi">10.1371/journal.pone.0343751.r003</article-id>
<title-group>
<article-title>Decision Letter 1</article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name name-style="western"><surname>Wroński</surname>
<given-names>Stanisław</given-names>
</name>
<role>Academic Editor</role>
</contrib>
<contrib contrib-type="author">
<name name-style="western"><surname>Wroński</surname>
<given-names>Stanisław</given-names>
</name>
<role>Academic Editor</role>
</contrib>
</contrib-group>
<permissions>
<copyright-year>2026</copyright-year>
<copyright-holder>Stanisław Wroński</copyright-holder>
<license xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">
<license-p>This is an open access article distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">Creative Commons Attribution License</ext-link>, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.</license-p></license>
</permissions>
<related-object document-id="10.1371/journal.pone.0343751" document-id-type="doi" document-type="article" id="rel-obj003" link-type="peer-reviewed-article"/>
<custom-meta-group>
<custom-meta>
<meta-name>Submission Version</meta-name>
<meta-value>1</meta-value>
</custom-meta>
</custom-meta-group>
</front-stub>
<body>
<p><named-content content-type="letter-date">29 Jun 2026</named-content></p>
<p>Joint modelling of PSA dynamics and prostate cancer risks: A population-based study in men without prior prostate cancer</p>
<p>PONE-D-26-05935R1</p>
<p>Dear Dr. B. Akynkozhayev</p>
<p>We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.</p>
<p>Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.</p>
<p>An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at <ext-link ext-link-type="uri" xlink:href="https://www.editorialmanager.com/pone/" xlink:type="simple">Editorial Manager®</ext-link> and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact <ext-link ext-link-type="uri" xlink:href="https://plos.my.site.com/s/" xlink:type="simple">billing support</ext-link>.</p>
<p>If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact <email xlink:type="simple">onepress@plos.org</email>.</p>
<p>Kind regards,</p>
<p>Stanisław Jacek Wroński, M.D., Ph.D, FEBU</p>
<p>Academic Editor</p>
<p>PLOS One</p>
<p>Reviewers' comments:</p>
<p>Reviewer's Responses to Questions</p>
<p>--&gt;<bold>Comments to the Author</bold></p>
<p>1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.--&gt;</p>
<p>Reviewer #2: All comments have been addressed</p>
<p>**********</p>
<p>--&gt;2. Is the manuscript technically sound, and do the data support the conclusions?</p>
<p>The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. --&gt;</p>
<p>Reviewer #2: (No Response)</p>
<p>**********</p>
<p>--&gt;3. Has the statistical analysis been performed appropriately and rigorously? --&gt;</p>
<p>Reviewer #2: (No Response)</p>
<p>**********</p>
<p>--&gt;4. Have the authors made all data underlying the findings in their manuscript fully available?</p>
<p>The <ext-link ext-link-type="uri" xlink:href="http://www.plosone.org/static/policies.action#sharing" xlink:type="simple">PLOS Data policy</ext-link> requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--&gt;</p>
<p>Reviewer #2: (No Response)</p>
<p>**********</p>
<p>--&gt;5. Is the manuscript presented in an intelligible fashion and written in standard English?</p>
<p>PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.--&gt;</p>
<p>Reviewer #2: (No Response)</p>
<p>**********</p>
<p>--&gt;6. Review Comments to the Author</p>
<p>Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)--&gt;</p>
<p>Reviewer #2: (No Response)</p>
<p>**********</p>
<p>--&gt;7. PLOS authors have the option to publish the peer review history of their article (<ext-link ext-link-type="uri" xlink:href="https://journals.plos.org/plosone/s/editorial-and-peer-review-process#loc-peer-review-history" xlink:type="simple">what does this mean?</ext-link>). If published, this will include your full peer review and any attached files.</p>
<p>If you choose “no”, your identity will remain anonymous but your review may still be made public.</p>
<p><bold>Do you want your identity to be public for this peer review?</bold>  For information about this choice, including consent withdrawal, please see our <ext-link ext-link-type="uri" xlink:href="https://www.plos.org/privacy-policy" xlink:type="simple">Privacy Policy</ext-link>.--&gt;</p>
<p>Reviewer #2: <bold>Yes:</bold> Rolf Gedeborg</p>
<p>**********</p>
</body>
</sub-article>
<sub-article article-type="editor-report" id="pone.0343751.r004" specific-use="acceptance-letter">
<front-stub>
<article-id pub-id-type="doi">10.1371/journal.pone.0343751.r004</article-id>
<title-group>
<article-title>Acceptance letter</article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name name-style="western"><surname>Wroński</surname>
<given-names>Stanisław</given-names>
</name>
<role>Academic Editor</role>
</contrib>
<contrib contrib-type="author">
<name name-style="western"><surname>Wroński</surname>
<given-names>Stanisław</given-names>
</name>
<role>Academic Editor</role>
</contrib>
</contrib-group>
<permissions>
<copyright-year>2026</copyright-year>
<copyright-holder>Stanisław Wroński</copyright-holder>
<license xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">
<license-p>This is an open access article distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/" xlink:type="simple">Creative Commons Attribution License</ext-link>, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.</license-p></license>
</permissions>
<related-object document-id="10.1371/journal.pone.0343751" document-id-type="doi" document-type="article" id="rel-obj004" link-type="peer-reviewed-article"/>
</front-stub>
<body>
<p>PONE-D-26-05935R1</p>
<p>PLOS One</p>
<p>Dear Dr. Akynkozhayev,</p>
<p>I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team.</p>
<p>At this stage, our production department will prepare your paper for publication. This includes ensuring the following:</p>
<p>* All references, tables, and figures are properly cited</p>
<p>* All relevant supporting information is included in the manuscript submission,</p>
<p>* There are no issues that prevent the paper from being properly typeset</p>
<p>You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps.</p>
<p>Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact <email xlink:type="simple">onepress@plos.org</email>.</p>
<p>You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at <ext-link ext-link-type="uri" xlink:href="https://explore.plos.org/phishing" xlink:type="simple">https://explore.plos.org/phishing</ext-link>.</p>
<p>If we can help with anything else, please email us at <email xlink:type="simple">customercare@plos.org</email>.</p>
<p>Thank you for submitting your work to PLOS ONE and supporting open access.</p>
<p>Kind regards,</p>
<p>PLOS ONE Editorial Office Staff</p>
<p>on behalf of</p>
<p>Dr. Stanisław Jacek Wroński</p>
<p>Academic Editor</p>
<p>PLOS One</p>
</body>
</sub-article>
</article>