Clinical Phenotypes of Interstitial Lung Disease Progression Identified by Longitudinal Quantitative CT: Support for an Inflammatory-Fibrotic Continuum

Clinical Phenotypes of Interstitial Lung Disease Progression Identified by Longitudinal Quantitative CT: Support for an Inflammatory-Fibrotic Continuum | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clinical Phenotypes of Interstitial Lung Disease Progression Identified by Longitudinal Quantitative CT: Support for an Inflammatory-Fibrotic Continuum Hugo Trabadelo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9237519/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Interstitial lung disease (ILD) is often managed as discrete diagnostic categories, yet many patients appear to evolve along a shared inflammatory-fibrotic continuum. Longitudinal quantitative CT enables objective characterization of disease trajectories through volumetric and densitometric metrics. Objectives: To identify and characterize clinically distinct progression phenotypes in ILD using longitudinal quantitative CT analysis. Methods: Retrospective cohort study of 16 patients with ILD undergoing serial chest CT (18 longitudinal comparisons, follow-up 0.5–19.6 months). Each comparison was quantified using a hybrid HU–Z-score method. Unsupervised K-means clustering (k=4) identified natural phenotypic groupings. Silhouette score assessed cluster quality. Results: Four clinically distinct progression phenotypes emerged: (1) Stable/Regressive (72%, n=13): mean volume change −12.5%, mean ΔZ-score −3.87; (2) True Progressive (17%, n=3): concurrent territorial expansion +68.7% and densification ΔZ +13.96, rate +2.37 Z-score units/month; (3) Hyperacute Crisis (6%, n=1): explosive progression +47.8% volume, ΔZ +22.56 over <1 month; (4) Contraction/Regression (6%, n=1): marked volume loss −38.9%, ΔZ −11.20. Within-patient trajectory analysis demonstrated phenotype transitions following treatment initiation, supporting non-static disease behavior. Clustering quality: silhouette score 0.322. Conclusions: Longitudinal quantitative CT identifies four clinically relevant progression phenotypes in ILD supporting a dynamic inflammatory-fibrotic continuum. The True Progressive Phenotype (17%) defines the clearest target for antifibrotic therapy, while the predominant Stable Phenotype (72%) supports de-escalation strategies. This objective phenotyping approach could enable earlier stage-appropriate intervention before irreversible fibrosis is established. Pulmonology interstitial lung disease inflammatory-fibrotic continuum quantitative CT disease progression phenotypes antifibrotic therapy INTRODUCTION Progressive pulmonary fibrosis (PPF) encompasses diverse interstitial lung diseases characterized by variable progression rates, treatment responses, and survival outcomes.[ 1 , 2 ] Conventional classification systems based on histology, radiology patterns, or single-timepoint pulmonary function thresholds capture baseline disease appearance but may not fully reflect the dynamic biology of progression.[ 3 , 4 ] An alternative interpretation is that many ILD presentations represent different positions along a shared inflammatory-fibrotic continuum — from early ground-glass and infiltrative abnormalities toward irreversible architectural fibrosis. Current treatment paradigms often intervene only after fibrosis is established and confirmed by functional decline, typically requiring ≥ 10% FVC decline or ≥ 15% DLCO decline over 12–24 months.[ 5 – 8 ] This retrospective approach may miss earlier therapeutic windows during predominantly inflammatory or mixed stages, when anti-inflammatory or immunosuppressive treatment might prevent irreversible fibrosis and reduce later dependence on costly antifibrotic therapy. Quantitative CT analysis enables objective, reproducible measurement of parenchymal involvement beyond visual scoring.[ 9 , 10 ] A hybrid quantification approach combining volumetric changes (territorial expansion or contraction) with densitometric changes (increased attenuation consistent with fibrosis) captures complementary dimensions of progression and generates datasets suitable for data-driven phenotype discovery.[ 11 ] We hypothesized that unsupervised analysis of longitudinal quantitative CT features would reveal clinically meaningful progression phenotypes aligned with an inflammatory-fibrotic continuum. Our objectives were: (1) identify distinct progression phenotypes, (2) characterize their quantitative and temporal profiles, and (3) propose stage-appropriate clinical implications for treatment decision-making. METHODS Study design and patients Retrospective cohort study of patients with ILD who underwent serial chest CT scans at our institution (Clínica San Bernardo, Buenos Aires, Argentina). Inclusion criteria: (1) diagnosis of ILD per consensus criteria;[ 12 ] (2) at least two chest CT scans separated by ≥ 0.5 months; (3) CT scans technically suitable for quantitative analysis. Exclusion criteria: predominant emphysema, pure consolidative process, or significant motion artifact. A total of 16 patients contributed 18 longitudinal CT comparisons. The study was conducted in accordance with the Declaration of Helsinki. Fully de-identified retrospective data were analyzed under written institutional authorization (Clínica San Bernardo, February 2026). Individual informed consent was waived per Argentine Resolución 1480/2011 for secondary use of anonymized data. Quantitative CT analysis — Hybrid HU–Z-score method Longitudinal CT quantification used the hybrid HU–Z-score method previously described by our group.[ 11 ] In brief: (1) automated lung parenchyma segmentation; (2) total lung volume (mL) measured at baseline and follow-up, with percentage volume change calculated; (3) Z-score quantifying the proportion of parenchyma exceeding normal attenuation thresholds (densitometric component); (4) derived hybrid metrics including rate of change per month, densification-to-volume ratios, and magnitude metrics. For each longitudinal comparison, 23 quantitative features were extracted encompassing temporal, volumetric, densitometric, and fibrosis-specific dimensions. Phenotype identification — Unsupervised clustering Unsupervised K-means clustering was applied to the standardized (mean = 0, SD = 1) feature matrix.[ 13 ] The choice of k = 4 was based on clinical interpretability. Silhouette score assessed cluster coherence and separation. Dimensionality reduction via Principal Component Analysis (PCA) and t-SNE[ 14 ] facilitated visualization. Within-patient trajectory analysis in patients with ≥ 3 serial scans examined phenotype transitions across consecutive intervals. All analyses were performed in Python (scikit-learn, pandas, matplotlib). RESULTS Cohort characteristics Sixteen patients with ILD contributed 18 longitudinal CT comparisons. Median follow-up was 12.5 months (range 0.5–19.6 months). Mean baseline Z-score was 25.8, mean baseline lung volume 6,384 mL, and mean baseline fibrotic percentage 19.8%. Two patients had ≥ 3 serial scans, contributing multiple consecutive pairwise comparisons. Four distinct progression phenotypes Unsupervised clustering identified four phenotypes with distinct quantitative profiles. Overall clustering quality: silhouette score 0.322, indicating moderate but distinguishable separation consistent with underlying biological overlap. Phenotype 1: Stable/Regressive (n = 13, 72%) Mean volume change − 12.5% (contraction), mean ΔZ-score − 3.87 (densitometric improvement), mean progression rate − 0.03 Z-score units/month. This predominant phenotype represents stable or improving disease. Volumetric contraction may reflect resolution of reversible inflammatory components (ground-glass opacities, organizing pneumonia patterns) responding to therapy. The high prevalence (72%) suggests that the majority of patients under follow-up at specialized ILD centers have non-progressive or regressive behavior, reflecting effective immunosuppression or antifibrotic therapy, spontaneous stabilization, or enrichment for less-aggressive ILD subtypes in ambulatory cohorts. Phenotype 2: True Progressive (n = 3, 17%) Mean volume change + 68.7% (massive territorial expansion), mean ΔZ-score + 13.96 (substantial densification), mean progression rate + 2.37 Z-score units/month, densification-to-volume ratio + 0.13. Concurrent territorial expansion and densification distinguish this phenotype from isolated volume changes (e.g., atelectasis). This pattern is the hallmark of true progressive pulmonary fibrosis and would be expected to meet conventional functional decline criteria (FVC/DLCO loss) over comparable timeframes. Despite comprising only 17% of observations, this group drives ILD morbidity and mortality. Phenotype 3: Hyperacute Crisis (n = 1, 6%) Volume change + 47.8%, ΔZ-score + 22.56, progression rate + 32.23 Z-score units/month over < 1 month — incompatible with chronic progression kinetics. This pattern is consistent with acute exacerbation of ILD or superimposed injury (organizing pneumonia, drug reaction, infection). Acute exacerbations of ILD carry mortality of 30–50%.[ 15 , 16 ] Based on a single observation; larger cohorts are needed to validate this as a robust phenotype vs. outlier. Phenotype 4: Contraction/Regression (n = 1, 6%) Volume change − 38.9%, ΔZ-score − 11.20, progression rate − 12.44 Z-score units/month. Rapid volume loss with densitometric improvement is inconsistent with progressive fibrosis and may represent treatment response of inflammatory/organizing components (especially in CTD-ILD or hypersensitivity pneumonitis after antigen removal[ 18 ]), spontaneous regression, or technical artifact. Clinical correlation is essential before interpreting as true disease regression. Based on a single observation. Within-patient trajectory dynamics Patient P002 contributed three consecutive comparisons over 28 months: Interval 1 (months 0–9.2) classified as True Progressive (volume + 70.1%, ΔZ + 8.22); Interval 2 (months 9.2–13.1) transitioned to Stable (volume + 5.1%, ΔZ − 1.14); Interval 3 (months 13.1–27.6) remained Stable (volume − 4.9%, ΔZ + 1.06). Antifibrotic therapy was initiated after Interval 1. This example demonstrates: (a) non-static phenotypes — patients transition between phenotypic states; (b) treatment-induced trajectory change; (c) the value of serial phenotyping to capture treatment response beyond single-timepoint assessment. PCA visualization Two-dimensional PCA projection captured 64% of feature variance (PC1: 42.1%, PC2: 21.9%). Phenotype clusters showed partial separation with overlap, consistent with continuous underlying biology rather than discrete disease entities. t-SNE visualization revealed local neighborhood structure with the True Progressive Phenotype forming a distinct cluster separate from the Stable Phenotype. DISCUSSION This study demonstrates that data-driven analysis of longitudinal quantitative CT identifies four clinically distinct progression phenotypes in ILD. The True Progressive Phenotype (17%), characterized by concurrent territorial expansion and densification, represents established fibrotic progression — the optimal target for antifibrotic therapy and clinical trial enrollment. The predominant Stable/Regressive Phenotype (72%) suggests that most patients under follow-up are non-progressive or treatment-responsive, supporting de-escalation strategies for those with significant treatment burdens. Within-patient trajectory analysis confirms that phenotypes are dynamic rather than fixed, with documented transitions after treatment initiation. Precision medicine and treatment stratification Current ILD paradigms often apply uniform strategies once fibrosis is recognized. Our phenotype-stratified continuum approach supports: (1) antifibrotic initiation or escalation for True Progressive patients; (2) observation or de-escalation for Stable/Regressive patients — avoiding the adverse effects (~ $ 90,000–100,000 USD/year cost, diarrhea, nausea) of antifibrotic therapy in non-progressors; (3) urgent evaluation for Hyperacute Crisis patterns; (4) continued current therapy with clinical correlation for Contraction patterns. Serial CT phenotyping identifies progression 3–6 months earlier than FVC-based criteria, potentially widening the therapeutic window before irreversible fibrosis. Comparison with existing approaches FVC-based progression criteria (≥ 10% decline over 12–24 months)[ 8 ] are retrospective, threshold-dependent, and slow — requiring over a year to confirm progression. CT phenotyping enables prospective classification at 3–6 months using continuous metrics without arbitrary thresholds. Traditional histologic/radiologic classifications (IPF vs. NSIP vs. COP) are static, based on baseline appearance; our approach is dynamic, capturing change over time and directly measuring progression rather than inferring prognosis from pattern. Combined with biomarkers (KL-6, MMP-7, CCL-18), CT phenotypes could enable multi-modal precision prognostication. Enrichment of clinical trials Heterogeneity in ILD trial populations — where many patients remain stable — dilutes treatment effect signals. Phenotype-based enrollment enriching for True Progressive patients could reduce required sample sizes, shorten trial duration, and improve cost-efficiency of antifibrotic drug development. Limitations Key limitations include: (1) small sample size (N = 16, 18 comparisons) — phenotypes with n = 1 should be interpreted as provisional rather than validated entities; (2) absence of complete pulmonary function (FVC, DLCO), specific ILD subtype, treatment, and hard outcome data for all patients; (3) single-center design with likely selection bias toward treated, monitored ILD patients; (4) variable follow-up intervals (0.5–19.6 months) affecting progression rate calculations; (5) K-means assumes spherical clusters and requires pre-specifying k — other algorithms may yield different structures; (6) segmentation errors or inspiratory effort variability could affect volumetric accuracy. Multi-center validation cohorts (N = 100–500) with standardized protocols, complete clinical data, and ≥ 24-month follow-up are needed before clinical deployment. CONCLUSIONS Longitudinal quantitative CT analysis identifies four clinically relevant progression phenotypes in ILD supporting a dynamic inflammatory-fibrotic continuum. The True Progressive Phenotype (17%) — concurrent territorial expansion and densification — is the optimal target for antifibrotic therapy. The predominant Stable/Regressive Phenotype (72%) supports observation and de-escalation strategies. Dynamic phenotype transitions following treatment confirm the non-static nature of ILD and the value of serial objective CT assessment. Larger prospective validation studies integrating clinical outcomes, biomarkers, and treatment exposure are needed to translate these findings into evidence-based treatment algorithms. Declarations AUTHOR CONTRIBUTIONS H. Trabadelo: concept and design, data acquisition, quantitative CT analysis, statistical analysis, manuscript writing and revision, final approval. ETHICS STATEMENT This retrospective observational study was conducted in accordance with the Declaration of Helsinki and Argentine national health research regulations (Resolución 1480/2011). Fully de-identified radiological data from routine clinical practice at Clínica San Bernardo, Buenos Aires, Argentina were analyzed under written institutional authorization (February 2026). Individual informed consent was waived per Argentine regulations for secondary analysis of anonymized data. No IRB/ethics committee review was required. All CT images were irreversibly de-identified prior to analysis. FUNDING AND CONFLICTS OF INTEREST This research received no external funding. The author declares no conflicts of interest. ACKNOWLEDGMENTS The author thanks the radiology and pulmonology teams at Clínica San Bernardo for clinical care of patients included in this study. References Raghu G, Remy-Jardin M, Myers JL, et al. Diagnosis of Idiopathic Pulmonary Fibrosis: An Official ATS/ERS/JRS/ALAT Clinical Practice Guideline. Am J Respir Crit Care Med. 2018;198(5):e44-e68. Cottin V, Hirani NA, Hotchkin DL, et al. Presentation, diagnosis and clinical course of the spectrum of progressive-fibrosing interstitial lung diseases. Eur Respir Rev. 2018;27(150):180076. Lynch DA, Sverzellati N, Travis WD, et al. Diagnostic criteria for idiopathic pulmonary fibrosis: a Fleischner Society White Paper. Lancet Respir Med. 2018;6(2):138-153. Richeldi L, Collard HR, Jones MG. Idiopathic pulmonary fibrosis. Lancet. 2017;389(10082):1941-1952. Richeldi L, du Bois RM, Raghu G, et al. Efficacy and safety of nintedanib in idiopathic pulmonary fibrosis. N Engl J Med. 2014;370(22):2071-2082. King TE Jr, Bradford WZ, Castro-Bernardini S, et al. A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N Engl J Med. 2014;370(22):2083-2092. Flaherty KR, Wells AU, Cottin V, et al. Nintedanib in Progressive Fibrosing Interstitial Lung Diseases. N Engl J Med. 2019;381(18):1718-1727. Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183(6):788-824. Jacob J, Bartholmai BJ, Rajagopalan S, et al. Mortality prediction in idiopathic pulmonary fibrosis: evaluation of computer-based CT analysis with conventional severity measures. Eur Respir J. 2017;49(1):1601011. Maldonado F, Moua T, Rajagopalan S, et al. Automated quantification of radiologic patterns predicts survival in idiopathic pulmonary fibrosis. Eur Respir J. 2014;43(1):204-212. Trabadelo H. Hybrid HU-Z-score longitudinal quantification for early fibrosis progression assessment. medRxiv. 2026. doi:10.64898/2026.03.03.26347353 American Thoracic Society; European Respiratory Society. ATS/ERS International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. Am J Respir Crit Care Med. 2002;165(2):277-304. MacQueen J. Some methods for classification and analysis of multivariate observations. Proc Fifth Berkeley Symp Math Stat Probab. 1967;1:281-297. van der Maaten L, Hinton G. Visualizing Data using t-SNE. J Mach Learn Res. 2008;9:2579-2605. Collard HR, Ryerson CJ, Corte TJ, et al. Acute Exacerbation of Idiopathic Pulmonary Fibrosis: An International Working Group Report. Am J Respir Crit Care Med. 2016;194(3):265-275. Kondoh Y, Taniguchi H, Katsuta T, et al. Risk factors of acute exacerbation of idiopathic pulmonary fibrosis. Sarcoidosis Vasc Diffuse Lung Dis. 2010;27(2):103-110. Kreuter M, Polke M, Walsh SLF, et al. Acute exacerbation of idiopathic pulmonary fibrosis: international survey and call for harmonisation. Eur Respir J. 2020;55(4):1901760. Fernandez Perez ER, Swigris JJ, Forssen AV, et al. Identifying an inciting antigen is associated with improved survival in patients with chronic hypersensitivity pneumonitis. Chest. 2013;144(5):1644-1651. Table TABLE 1. Quantitative profile of the four progression phenotypes Stable/Regressive (n=13, 72%) True Progressive (n=3, 17%) Hyperacute Crisis (n=1, 6%) Contraction (n=1, 6%) Volume change (%) −12.5 +68.7 +47.8 −38.9 ΔZ-score −3.87 +13.96 +22.56 −11.20 Rate (ΔZ/month) −0.03 +2.37 +32.23 −12.44 Key pattern Stable/improving Expansion + densification Explosive (<1 mo) Rapid regression Treatment priority Low / de-escalate HIGH — antifibrotic Urgent evaluation Continue; correlate Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9237519","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":612871442,"identity":"de3404a8-f251-4742-a884-95b8ad0277fe","order_by":0,"name":"Hugo Trabadelo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIiWNgGAWjYBACAyBmBiE+BuYDQLaEDGEtbFAtbAxsCSAtPKRo4QHZyEBYi7l878PPBTXWcmzsPZ9f3aix4GFgP3x0Az4tlm3sxtIzjqUbs/Gc3WadcwzoMJ60tBt4HXaMjUGah+1wYptE7jbjHDagFgkeM0JamH/z/Dtc3yaR88w45x9xWtikedsOJ7BJ5DA/zm0jSksam/XMvnTDNp5jZsy5fRI8bAT9cvgY8+2Cb9by/OzNjz/nfKuT42c/fAyvFmTAJgEmiVUOAswfSFE9CkbBKBgFIwcAAE48Po8Nud2cAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0006-3817-9693","institution":"Clinica San Bernardo","correspondingAuthor":true,"prefix":"","firstName":"Hugo","middleName":"","lastName":"Trabadelo","suffix":""}],"badges":[],"createdAt":"2026-03-26 19:14:47","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9237519/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9237519/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105903960,"identity":"2fbfa6a8-c8c0-41b4-a716-1f9be2596a80","added_by":"auto","created_at":"2026-04-01 09:59:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":664841,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9237519/v1/bb65c213-b985-4bd4-b923-2c846a57b7f5.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eClinical Phenotypes of Interstitial Lung Disease Progression Identified by Longitudinal Quantitative CT: Support for an Inflammatory-Fibrotic Continuum\u003c/p\u003e","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eProgressive pulmonary fibrosis (PPF) encompasses diverse interstitial lung diseases characterized by variable progression rates, treatment responses, and survival outcomes.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] Conventional classification systems based on histology, radiology patterns, or single-timepoint pulmonary function thresholds capture baseline disease appearance but may not fully reflect the dynamic biology of progression.[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] An alternative interpretation is that many ILD presentations represent different positions along a shared inflammatory-fibrotic continuum \u0026mdash; from early ground-glass and infiltrative abnormalities toward irreversible architectural fibrosis.\u003c/p\u003e \u003cp\u003eCurrent treatment paradigms often intervene only after fibrosis is established and confirmed by functional decline, typically requiring\u0026thinsp;\u0026ge;\u0026thinsp;10% FVC decline or \u0026ge;\u0026thinsp;15% DLCO decline over 12\u0026ndash;24 months.[\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] This retrospective approach may miss earlier therapeutic windows during predominantly inflammatory or mixed stages, when anti-inflammatory or immunosuppressive treatment might prevent irreversible fibrosis and reduce later dependence on costly antifibrotic therapy.\u003c/p\u003e \u003cp\u003eQuantitative CT analysis enables objective, reproducible measurement of parenchymal involvement beyond visual scoring.[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] A hybrid quantification approach combining volumetric changes (territorial expansion or contraction) with densitometric changes (increased attenuation consistent with fibrosis) captures complementary dimensions of progression and generates datasets suitable for data-driven phenotype discovery.[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eWe hypothesized that unsupervised analysis of longitudinal quantitative CT features would reveal clinically meaningful progression phenotypes aligned with an inflammatory-fibrotic continuum. Our objectives were: (1) identify distinct progression phenotypes, (2) characterize their quantitative and temporal profiles, and (3) propose stage-appropriate clinical implications for treatment decision-making.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design and patients\u003c/h2\u003e \u003cp\u003eRetrospective cohort study of patients with ILD who underwent serial chest CT scans at our institution (Cl\u0026iacute;nica San Bernardo, Buenos Aires, Argentina). Inclusion criteria: (1) diagnosis of ILD per consensus criteria;[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] (2) at least two chest CT scans separated by \u0026ge;\u0026thinsp;0.5 months; (3) CT scans technically suitable for quantitative analysis. Exclusion criteria: predominant emphysema, pure consolidative process, or significant motion artifact. A total of 16 patients contributed 18 longitudinal CT comparisons. The study was conducted in accordance with the Declaration of Helsinki. Fully de-identified retrospective data were analyzed under written institutional authorization (Cl\u0026iacute;nica San Bernardo, February 2026). Individual informed consent was waived per Argentine Resoluci\u0026oacute;n 1480/2011 for secondary use of anonymized data.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eQuantitative CT analysis — Hybrid HU–Z-score method\u003c/h3\u003e\n\u003cp\u003eLongitudinal CT quantification used the hybrid HU\u0026ndash;Z-score method previously described by our group.[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] In brief: (1) automated lung parenchyma segmentation; (2) total lung volume (mL) measured at baseline and follow-up, with percentage volume change calculated; (3) Z-score quantifying the proportion of parenchyma exceeding normal attenuation thresholds (densitometric component); (4) derived hybrid metrics including rate of change per month, densification-to-volume ratios, and magnitude metrics. For each longitudinal comparison, 23 quantitative features were extracted encompassing temporal, volumetric, densitometric, and fibrosis-specific dimensions.\u003c/p\u003e\n\u003ch3\u003ePhenotype identification — Unsupervised clustering\u003c/h3\u003e\n\u003cp\u003eUnsupervised K-means clustering was applied to the standardized (mean\u0026thinsp;=\u0026thinsp;0, SD\u0026thinsp;=\u0026thinsp;1) feature matrix.[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] The choice of k\u0026thinsp;=\u0026thinsp;4 was based on clinical interpretability. Silhouette score assessed cluster coherence and separation. Dimensionality reduction via Principal Component Analysis (PCA) and t-SNE[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] facilitated visualization. Within-patient trajectory analysis in patients with \u0026ge;\u0026thinsp;3 serial scans examined phenotype transitions across consecutive intervals. All analyses were performed in Python (scikit-learn, pandas, matplotlib).\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eCohort characteristics\u003c/h2\u003e \u003cp\u003eSixteen patients with ILD contributed 18 longitudinal CT comparisons. Median follow-up was 12.5 months (range 0.5\u0026ndash;19.6 months). Mean baseline Z-score was 25.8, mean baseline lung volume 6,384 mL, and mean baseline fibrotic percentage 19.8%. Two patients had\u0026thinsp;\u0026ge;\u0026thinsp;3 serial scans, contributing multiple consecutive pairwise comparisons.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFour distinct progression phenotypes\u003c/h2\u003e \u003cp\u003eUnsupervised clustering identified four phenotypes with distinct quantitative profiles. Overall clustering quality: silhouette score 0.322, indicating moderate but distinguishable separation consistent with underlying biological overlap.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePhenotype 1: Stable/Regressive (n = 13, 72%)\u003c/h3\u003e\n\u003cp\u003eMean volume change\u0026thinsp;\u0026minus;\u0026thinsp;12.5% (contraction), mean ΔZ-score\u0026thinsp;\u0026minus;\u0026thinsp;3.87 (densitometric improvement), mean progression rate\u0026thinsp;\u0026minus;\u0026thinsp;0.03 Z-score units/month. This predominant phenotype represents stable or improving disease. Volumetric contraction may reflect resolution of reversible inflammatory components (ground-glass opacities, organizing pneumonia patterns) responding to therapy. The high prevalence (72%) suggests that the majority of patients under follow-up at specialized ILD centers have non-progressive or regressive behavior, reflecting effective immunosuppression or antifibrotic therapy, spontaneous stabilization, or enrichment for less-aggressive ILD subtypes in ambulatory cohorts.\u003c/p\u003e\n\u003ch3\u003ePhenotype 2: True Progressive (n = 3, 17%)\u003c/h3\u003e\n\u003cp\u003eMean volume change\u0026thinsp;+\u0026thinsp;68.7% (massive territorial expansion), mean ΔZ-score\u0026thinsp;+\u0026thinsp;13.96 (substantial densification), mean progression rate\u0026thinsp;+\u0026thinsp;2.37 Z-score units/month, densification-to-volume ratio\u0026thinsp;+\u0026thinsp;0.13. Concurrent territorial expansion and densification distinguish this phenotype from isolated volume changes (e.g., atelectasis). This pattern is the hallmark of true progressive pulmonary fibrosis and would be expected to meet conventional functional decline criteria (FVC/DLCO loss) over comparable timeframes. Despite comprising only 17% of observations, this group drives ILD morbidity and mortality.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePhenotype 3: Hyperacute Crisis (n\u0026thinsp;=\u0026thinsp;1, 6%)\u003c/h2\u003e \u003cp\u003eVolume change\u0026thinsp;+\u0026thinsp;47.8%, ΔZ-score\u0026thinsp;+\u0026thinsp;22.56, progression rate\u0026thinsp;+\u0026thinsp;32.23 Z-score units/month over \u0026lt;\u0026thinsp;1 month \u0026mdash; incompatible with chronic progression kinetics. This pattern is consistent with acute exacerbation of ILD or superimposed injury (organizing pneumonia, drug reaction, infection). Acute exacerbations of ILD carry mortality of 30\u0026ndash;50%.[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] Based on a single observation; larger cohorts are needed to validate this as a robust phenotype vs. outlier.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003ePhenotype 4: Contraction/Regression (n\u0026thinsp;=\u0026thinsp;1, 6%)\u003c/h2\u003e \u003cp\u003eVolume change\u0026thinsp;\u0026minus;\u0026thinsp;38.9%, ΔZ-score\u0026thinsp;\u0026minus;\u0026thinsp;11.20, progression rate\u0026thinsp;\u0026minus;\u0026thinsp;12.44 Z-score units/month. Rapid volume loss with densitometric improvement is inconsistent with progressive fibrosis and may represent treatment response of inflammatory/organizing components (especially in CTD-ILD or hypersensitivity pneumonitis after antigen removal[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]), spontaneous regression, or technical artifact. Clinical correlation is essential before interpreting as true disease regression. Based on a single observation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eWithin-patient trajectory dynamics\u003c/h2\u003e \u003cp\u003ePatient P002 contributed three consecutive comparisons over 28 months: Interval 1 (months 0\u0026ndash;9.2) classified as True Progressive (volume\u0026thinsp;+\u0026thinsp;70.1%, ΔZ\u0026thinsp;+\u0026thinsp;8.22); Interval 2 (months 9.2\u0026ndash;13.1) transitioned to Stable (volume\u0026thinsp;+\u0026thinsp;5.1%, ΔZ\u0026thinsp;\u0026minus;\u0026thinsp;1.14); Interval 3 (months 13.1\u0026ndash;27.6) remained Stable (volume\u0026thinsp;\u0026minus;\u0026thinsp;4.9%, ΔZ\u0026thinsp;+\u0026thinsp;1.06). Antifibrotic therapy was initiated after Interval 1. This example demonstrates: (a) non-static phenotypes \u0026mdash; patients transition between phenotypic states; (b) treatment-induced trajectory change; (c) the value of serial phenotyping to capture treatment response beyond single-timepoint assessment.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003ePCA visualization\u003c/h2\u003e \u003cp\u003eTwo-dimensional PCA projection captured 64% of feature variance (PC1: 42.1%, PC2: 21.9%). Phenotype clusters showed partial separation with overlap, consistent with continuous underlying biology rather than discrete disease entities. t-SNE visualization revealed local neighborhood structure with the True Progressive Phenotype forming a distinct cluster separate from the Stable Phenotype.\u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study demonstrates that data-driven analysis of longitudinal quantitative CT identifies four clinically distinct progression phenotypes in ILD. The True Progressive Phenotype (17%), characterized by concurrent territorial expansion and densification, represents established fibrotic progression \u0026mdash; the optimal target for antifibrotic therapy and clinical trial enrollment. The predominant Stable/Regressive Phenotype (72%) suggests that most patients under follow-up are non-progressive or treatment-responsive, supporting de-escalation strategies for those with significant treatment burdens. Within-patient trajectory analysis confirms that phenotypes are dynamic rather than fixed, with documented transitions after treatment initiation.\u003c/p\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003ePrecision medicine and treatment stratification\u003c/h2\u003e \u003cp\u003eCurrent ILD paradigms often apply uniform strategies once fibrosis is recognized. Our phenotype-stratified continuum approach supports: (1) antifibrotic initiation or escalation for True Progressive patients; (2) observation or de-escalation for Stable/Regressive patients \u0026mdash; avoiding the adverse effects (~\u003cspan\u003e$\u003c/span\u003e90,000\u0026ndash;100,000 USD/year cost, diarrhea, nausea) of antifibrotic therapy in non-progressors; (3) urgent evaluation for Hyperacute Crisis patterns; (4) continued current therapy with clinical correlation for Contraction patterns. Serial CT phenotyping identifies progression 3\u0026ndash;6 months earlier than FVC-based criteria, potentially widening the therapeutic window before irreversible fibrosis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eComparison with existing approaches\u003c/h2\u003e \u003cp\u003eFVC-based progression criteria (\u0026ge;\u0026thinsp;10% decline over 12\u0026ndash;24 months)[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] are retrospective, threshold-dependent, and slow \u0026mdash; requiring over a year to confirm progression. CT phenotyping enables prospective classification at 3\u0026ndash;6 months using continuous metrics without arbitrary thresholds. Traditional histologic/radiologic classifications (IPF vs. NSIP vs. COP) are static, based on baseline appearance; our approach is dynamic, capturing change over time and directly measuring progression rather than inferring prognosis from pattern. Combined with biomarkers (KL-6, MMP-7, CCL-18), CT phenotypes could enable multi-modal precision prognostication.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eEnrichment of clinical trials\u003c/h2\u003e \u003cp\u003eHeterogeneity in ILD trial populations \u0026mdash; where many patients remain stable \u0026mdash; dilutes treatment effect signals. Phenotype-based enrollment enriching for True Progressive patients could reduce required sample sizes, shorten trial duration, and improve cost-efficiency of antifibrotic drug development.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eKey limitations include: (1) small sample size (N\u0026thinsp;=\u0026thinsp;16, 18 comparisons) \u0026mdash; phenotypes with n\u0026thinsp;=\u0026thinsp;1 should be interpreted as provisional rather than validated entities; (2) absence of complete pulmonary function (FVC, DLCO), specific ILD subtype, treatment, and hard outcome data for all patients; (3) single-center design with likely selection bias toward treated, monitored ILD patients; (4) variable follow-up intervals (0.5\u0026ndash;19.6 months) affecting progression rate calculations; (5) K-means assumes spherical clusters and requires pre-specifying k \u0026mdash; other algorithms may yield different structures; (6) segmentation errors or inspiratory effort variability could affect volumetric accuracy. Multi-center validation cohorts (N\u0026thinsp;=\u0026thinsp;100\u0026ndash;500) with standardized protocols, complete clinical data, and \u0026ge;\u0026thinsp;24-month follow-up are needed before clinical deployment.\u003c/p\u003e \u003c/div\u003e"},{"header":"CONCLUSIONS","content":"\u003cp\u003eLongitudinal quantitative CT analysis identifies four clinically relevant progression phenotypes in ILD supporting a dynamic inflammatory-fibrotic continuum. The True Progressive Phenotype (17%) \u0026mdash; concurrent territorial expansion and densification \u0026mdash; is the optimal target for antifibrotic therapy. The predominant Stable/Regressive Phenotype (72%) supports observation and de-escalation strategies. Dynamic phenotype transitions following treatment confirm the non-static nature of ILD and the value of serial objective CT assessment. Larger prospective validation studies integrating clinical outcomes, biomarkers, and treatment exposure are needed to translate these findings into evidence-based treatment algorithms.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAUTHOR CONTRIBUTIONS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eH. Trabadelo: concept and design, data acquisition, quantitative CT analysis, statistical analysis, manuscript writing and revision, final approval.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eETHICS STATEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective observational study was conducted in accordance with the Declaration of Helsinki and Argentine national health research regulations (Resolución 1480/2011). Fully de-identified radiological data from routine clinical practice at Clínica San Bernardo, Buenos Aires, Argentina were analyzed under written institutional authorization (February 2026). Individual informed consent was waived per Argentine regulations for secondary analysis of anonymized data. No IRB/ethics committee review was required. All CT images were irreversibly de-identified prior to analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFUNDING AND CONFLICTS OF INTEREST\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no external funding. The author declares no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eACKNOWLEDGMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author thanks the radiology and pulmonology teams at Clínica San Bernardo for clinical care of patients included in this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eRaghu G, Remy-Jardin M, Myers JL, et al. Diagnosis of Idiopathic Pulmonary Fibrosis: An Official ATS/ERS/JRS/ALAT Clinical Practice Guideline. Am J Respir Crit Care Med. 2018;198(5):e44-e68.\u003c/li\u003e\n \u003cli\u003eCottin V, Hirani NA, Hotchkin DL, et al. Presentation, diagnosis and clinical course of the spectrum of progressive-fibrosing interstitial lung diseases. Eur Respir Rev. 2018;27(150):180076.\u003c/li\u003e\n \u003cli\u003eLynch DA, Sverzellati N, Travis WD, et al. Diagnostic criteria for idiopathic pulmonary fibrosis: a Fleischner Society White Paper. Lancet Respir Med. 2018;6(2):138-153.\u003c/li\u003e\n \u003cli\u003eRicheldi L, Collard HR, Jones MG. Idiopathic pulmonary fibrosis. Lancet. 2017;389(10082):1941-1952.\u003c/li\u003e\n \u003cli\u003eRicheldi L, du Bois RM, Raghu G, et al. Efficacy and safety of nintedanib in idiopathic pulmonary fibrosis. N Engl J Med. 2014;370(22):2071-2082.\u003c/li\u003e\n \u003cli\u003eKing TE Jr, Bradford WZ, Castro-Bernardini S, et al. A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N Engl J Med. 2014;370(22):2083-2092.\u003c/li\u003e\n \u003cli\u003eFlaherty KR, Wells AU, Cottin V, et al. Nintedanib in Progressive Fibrosing Interstitial Lung Diseases. N Engl J Med. 2019;381(18):1718-1727.\u003c/li\u003e\n \u003cli\u003eRaghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183(6):788-824.\u003c/li\u003e\n \u003cli\u003eJacob J, Bartholmai BJ, Rajagopalan S, et al. Mortality prediction in idiopathic pulmonary fibrosis: evaluation of computer-based CT analysis with conventional severity measures. Eur Respir J. 2017;49(1):1601011.\u003c/li\u003e\n \u003cli\u003eMaldonado F, Moua T, Rajagopalan S, et al. Automated quantification of radiologic patterns predicts survival in idiopathic pulmonary fibrosis. Eur Respir J. 2014;43(1):204-212.\u003c/li\u003e\n \u003cli\u003eTrabadelo H. Hybrid HU-Z-score longitudinal quantification for early fibrosis progression assessment. medRxiv. 2026. doi:10.64898/2026.03.03.26347353\u003c/li\u003e\n \u003cli\u003eAmerican Thoracic Society; European Respiratory Society. ATS/ERS International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. Am J Respir Crit Care Med. 2002;165(2):277-304.\u003c/li\u003e\n \u003cli\u003eMacQueen J. Some methods for classification and analysis of multivariate observations. Proc Fifth Berkeley Symp Math Stat Probab. 1967;1:281-297.\u003c/li\u003e\n \u003cli\u003evan der Maaten L, Hinton G. Visualizing Data using t-SNE. J Mach Learn Res. 2008;9:2579-2605.\u003c/li\u003e\n \u003cli\u003eCollard HR, Ryerson CJ, Corte TJ, et al. Acute Exacerbation of Idiopathic Pulmonary Fibrosis: An International Working Group Report. Am J Respir Crit Care Med. 2016;194(3):265-275.\u003c/li\u003e\n \u003cli\u003eKondoh Y, Taniguchi H, Katsuta T, et al. Risk factors of acute exacerbation of idiopathic pulmonary fibrosis. Sarcoidosis Vasc Diffuse Lung Dis. 2010;27(2):103-110.\u003c/li\u003e\n \u003cli\u003eKreuter M, Polke M, Walsh SLF, et al. Acute exacerbation of idiopathic pulmonary fibrosis: international survey and call for harmonisation. Eur Respir J. 2020;55(4):1901760.\u003c/li\u003e\n \u003cli\u003eFernandez Perez ER, Swigris JJ, Forssen AV, et al. Identifying an inciting antigen is associated with improved survival in patients with chronic hypersensitivity pneumonitis. Chest. 2013;144(5):1644-1651.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003e\u003cstrong\u003eTABLE 1. Quantitative profile of the four progression phenotypes\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStable/Regressive\u003cbr\u003e\u0026nbsp;(n=13, 72%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrue Progressive\u003cbr\u003e\u0026nbsp;(n=3, 17%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHyperacute Crisis\u003cbr\u003e\u0026nbsp;(n=1, 6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eContraction\u003cbr\u003e\u0026nbsp;(n=1, 6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eVolume change (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026minus;12.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e+68.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e+47.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026minus;38.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026Delta;Z-score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026minus;3.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e+13.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e+22.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026minus;11.20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eRate (\u0026Delta;Z/month)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026minus;0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e+2.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e+32.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e\u0026minus;12.44\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eKey pattern\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eStable/improving\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eExpansion + densification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eExplosive (\u0026lt;1 mo)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eRapid regression\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eTreatment priority\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eLow / de-escalate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eHIGH \u0026mdash; antifibrotic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eUrgent evaluation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eContinue; correlate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"interstitial lung disease, inflammatory-fibrotic continuum, quantitative CT, disease progression, phenotypes, antifibrotic therapy","lastPublishedDoi":"10.21203/rs.3.rs-9237519/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9237519/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eInterstitial lung disease (ILD) is often managed as discrete diagnostic categories, yet many patients appear to evolve along a shared inflammatory-fibrotic continuum. Longitudinal quantitative CT enables objective characterization of disease trajectories through volumetric and densitometric metrics.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjectives: \u003c/strong\u003eTo identify and characterize clinically distinct progression phenotypes in ILD using longitudinal quantitative CT analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eRetrospective cohort study of 16 patients with ILD undergoing serial chest CT (18 longitudinal comparisons, follow-up 0.5–19.6 months). Each comparison was quantified using a hybrid HU–Z-score method. Unsupervised K-means clustering (k=4) identified natural phenotypic groupings. Silhouette score assessed cluster quality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eFour clinically distinct progression phenotypes emerged: (1) Stable/Regressive (72%, n=13): mean volume change −12.5%, mean ΔZ-score −3.87; (2) True Progressive (17%, n=3): concurrent territorial expansion +68.7% and densification ΔZ +13.96, rate +2.37 Z-score units/month; (3) Hyperacute Crisis (6%, n=1): explosive progression +47.8% volume, ΔZ +22.56 over \u0026lt;1 month; (4) Contraction/Regression (6%, n=1): marked volume loss −38.9%, ΔZ −11.20. Within-patient trajectory analysis demonstrated phenotype transitions following treatment initiation, supporting non-static disease behavior. Clustering quality: silhouette score 0.322.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eLongitudinal quantitative CT identifies four clinically relevant progression phenotypes in ILD supporting a dynamic inflammatory-fibrotic continuum. The True Progressive Phenotype (17%) defines the clearest target for antifibrotic therapy, while the predominant Stable Phenotype (72%) supports de-escalation strategies. This objective phenotyping approach could enable earlier stage-appropriate intervention before irreversible fibrosis is established.\u003c/p\u003e","manuscriptTitle":"Clinical Phenotypes of Interstitial Lung Disease Progression Identified by Longitudinal Quantitative CT: Support for an Inflammatory-Fibrotic Continuum","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-30 04:56:06","doi":"10.21203/rs.3.rs-9237519/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5adbc555-d1c1-4ffe-af66-c60610041cbd","owner":[],"postedDate":"March 30th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65216671,"name":"Pulmonology"}],"tags":[],"updatedAt":"2026-03-30T04:56:06+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-30 04:56:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9237519","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9237519","identity":"rs-9237519","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}