Section Structure
This article is organized into the following sections: clinical background and unmet need; study design and methods; key quantitative results; clinical interpretation and translational relevance; strengths and limitations; implications for practice and research; and references.
Highlights
In idiopathic pulmonary fibrosis (IPF), 1-year change in quantitative CT-derived fibrosis score was independently associated with transplant-free survival across discovery and external validation cohorts.
Anchor-based minimal clinically important difference estimates for annual fibrosis score progression were 2.72% when anchored to forced vital capacity change and 4.52% when anchored to diffusing capacity change.
A prognostically optimized threshold of fibrosis score increase of at least 4.05% showed the most consistent discrimination of overall and 3-year transplant-free survival in external validation.
The study supports the concept that annual CT, interpreted quantitatively rather than visually alone, may help monitor disease trajectory and refine risk stratification in IPF.
Background
Idiopathic pulmonary fibrosis is a chronic, progressive fibrosing interstitial lung disease characterized by worsening respiratory symptoms, declining lung function, and premature mortality. Despite the availability of antifibrotic therapies, disease trajectories remain heterogeneous. Some patients decline steadily, while others experience relatively slow progression or abrupt deterioration. This variability creates a central challenge in IPF management: clinicians need reliable tools to determine whether disease is truly progressing, whether therapy is sufficient, and whether patients should be referred for transplant evaluation or intensification of supportive care.
Serial pulmonary function testing, particularly forced vital capacity (FVC), has long served as a cornerstone for monitoring IPF. Decline in FVC is clinically intuitive, prognostically meaningful, and embedded in trial design. Yet FVC has important limitations. It may not fully capture regional fibrotic progression, can be influenced by patient effort and comorbidity, and may miss structurally important changes that occur before substantial physiologic decline becomes evident. Diffusing capacity for carbon monoxide (DLco) adds complementary information but is also variable and can be confounded by pulmonary vascular disease, emphysema, and anemia.
High-resolution computed tomography is central to diagnosis and often used qualitatively during follow-up, but visual estimation of progression is observer-dependent. Quantitative CT (QCT) offers a more reproducible way to measure disease burden, including fibrosis extent, through automated or semi-automated image analysis. The unresolved question has not been whether QCT can detect progression, but rather what degree of change should be considered clinically meaningful. Minimal clinically important difference thresholds and prognostically useful cut points have not been well standardized across software platforms and algorithms.
The present study by Kim and colleagues addresses that gap by evaluating clinically meaningful and prognostically relevant thresholds for 1-year change in QCT-derived fibrosis score in IPF using a fully automated approach, and by testing those thresholds in an external validation cohort.
Study Design and Methods
Design and population
This was a multicenter retrospective study of patients with IPF who underwent both baseline and 1-year follow-up chest CT. The investigators assembled a discovery cohort of 524 patients and an external validation cohort of 224 patients. Mean age was 66.8 years in the discovery cohort and 69.6 years in the validation cohort; men comprised 79% and 83% of the cohorts, respectively. These demographics are broadly consistent with the epidemiology of IPF in specialist practice.
Quantitative imaging metric
The main imaging biomarker was 1-year change in fibrosis score, denoted ΔFS, derived from fully automated CT quantification. While the abstract does not detail every image-processing step, the conceptual endpoint is straightforward: a percentage-point change in CT-defined fibrosis extent between baseline and approximately 1 year.
Anchor-based assessment of clinical meaning
To estimate a minimal clinically important difference, the investigators used an anchor-based approach. This method links change in the imaging biomarker to changes in external clinical measures that clinicians already recognize as meaningful. Here, the anchors were 1-year changes in FVC and DLco. This is an important methodological choice because it grounds the imaging threshold in patient-relevant disease progression rather than relying only on statistical distribution characteristics.
Prognostic threshold derivation
Beyond MCID, the authors derived a prognostic threshold using maximally selected log-rank statistics. In practical terms, this approach identifies the cut point in ΔFS that best separates patients according to transplant-free survival. The resulting threshold was then tested in both the discovery and validation cohorts.
Clinical outcome
The principal outcome was transplant-free survival, a clinically meaningful composite endpoint in IPF because it captures both death and progression to lung transplantation. The study also examined 3-year transplant-free survival in validation analyses.
Key Findings
Meaningful thresholds for annual change in fibrosis score
The anchor-based MCID estimates differed slightly according to the physiologic anchor used. When linked to FVC change, the MCID for 1-year ΔFS was 2.72%. When linked to DLco change, the corresponding estimate was 4.52%. These values suggest that relatively modest increases in quantitative fibrosis extent over a year may be clinically meaningful, but that the exact threshold depends on which aspect of physiologic deterioration is used as reference.
The prognostically optimized threshold was 4.05%. This value sits between the two anchor-based estimates and appears especially attractive because it was derived from survival separation and then externally validated.
Association with transplant-free survival in the discovery cohort
In the discovery cohort, 1-year ΔFS was independently associated with transplant-free survival. Although the abstract does not report the exact adjusted hazard ratio for all analyses in the derivation set, it states that higher ΔFS conferred greater risk and that exceeding either MCID threshold or the prognostic threshold was significantly associated with worse transplant-free survival, with all P values below .001. This indicates that the relationship between structural progression on CT and hard clinical outcomes is robust and not merely a reflection of cross-sectional disease severity.
External validation confirms prognostic value
The external validation cohort is a major strength of the report. In this cohort, ΔFS remained significantly associated with transplant-free survival after adjustment, with an adjusted hazard ratio of 1.11 per unit increase in ΔFS and a 95% confidence interval of 1.04 to 1.18. This suggests that each incremental increase in annual fibrosis progression on CT carried additional risk.
All prespecified ΔFS thresholds were prognostic for overall transplant-free survival in validation analyses. However, the threshold of ΔFS at or above 4.05% emerged as the most consistently informative. For overall transplant-free survival, this threshold was associated with an adjusted hazard ratio of 2.78 (95% CI, 1.36-5.68). For 3-year transplant-free survival, the adjusted hazard ratio was 2.88 (95% CI, 1.11-7.48). These effect sizes are clinically meaningful and suggest that crossing the 4% annual progression mark identifies a substantially higher-risk phenotype.
Why the 4% threshold matters
From a clinical standpoint, a threshold near 4% has practical appeal. It is large enough to reduce the chance that trivial technical differences between scans are overinterpreted, yet small enough to identify progression before advanced functional decline becomes unavoidable. The consistency of the 4.05% threshold across overall and shorter-term survival analyses also strengthens its candidacy as an operational decision point for annual follow-up.
Clinical Interpretation
QCT as a complement, not a replacement, for pulmonary function
This study should not be interpreted as displacing FVC or DLco. Rather, it supports a more integrated monitoring framework in which quantitative structural progression on CT complements physiologic decline. A patient whose FVC appears relatively stable but whose fibrosis score increases by 4% or more over 1 year may warrant closer review for occult progression, treatment adherence, alternative explanations such as superimposed complications, or earlier transplant discussion.
Conversely, the study helps interpret situations in which PFT changes are equivocal. Day-to-day variability in FVC can complicate clinical decisions, especially when measured changes are modest. In such cases, objective evidence of worsening fibrosis on CT may increase confidence that disease progression is real rather than noise.
Potential role in longitudinal risk stratification
Risk stratification in IPF is evolving from static baseline assessment to dynamic longitudinal assessment. Baseline variables such as age, sex, FVC, DLco, oxygen requirement, and radiologic extent remain important, but change over time often conveys more prognostic information than a single measurement. This study fits squarely within that shift. A validated annual imaging threshold may help classify patients into lower- and higher-risk trajectories, potentially informing clinic visit intensity, pulmonary rehabilitation timing, palliative care integration, and transplant referral thresholds.
Implications for imaging follow-up strategy
The authors conclude that the data support annual CT follow-up with predefined QCT thresholds. That is a provocative and clinically relevant suggestion. In many centers, CT is repeated selectively because of cost, access, and cumulative radiation concerns. The study provides evidence that annual CT, if analyzed quantitatively and consistently, may provide actionable prognostic information rather than simply descriptive documentation of progression.
Still, implementation should be thoughtful. The net value of routine annual CT will depend on scanner standardization, availability of validated software, radiation-dose optimization, and how often imaging findings change management beyond what is already evident from symptoms and pulmonary function.
Strengths of the Study
Several features make this study important. First, the sample size was substantial for a CT-based IPF analysis, particularly with paired baseline and 1-year imaging. Second, the multicenter design improves relevance beyond a single institutional workflow. Third, the use of anchor-based MCID estimation is methodologically sound because it ties imaging change to clinically meaningful physiologic benchmarks. Fourth, the external validation cohort meaningfully strengthens credibility, especially because the main prognostic threshold remained associated with both overall and 3-year transplant-free survival.
Another strength is the focus on a fully automated CT-derived fibrosis metric. If such tools can be standardized across institutions, they may reduce reader variability and make quantitative imaging more scalable in both clinical practice and trials.
Limitations and Cautions
As with any retrospective imaging study, several limitations deserve attention. The first is generalizability across CT acquisition protocols and software environments. QCT biomarkers are vulnerable to technical variation including slice thickness, reconstruction kernel, inspiratory effort, and scanner platform. A threshold derived using one algorithm may not transfer perfectly to another, even if the conceptual metric is similar.
Second, the abstract does not provide granular information on antifibrotic treatment exposure, adherence, or treatment changes over the observation interval. Because antifibrotic therapy can slow disease progression, treatment heterogeneity could influence both ΔFS and survival relationships.
Third, transplant-free survival is clinically meaningful but can be influenced by local transplant referral and listing practices. In multicenter retrospective studies, this may introduce some variability unrelated to the biology of fibrosis progression itself.
Fourth, annual CT follow-up is not without tradeoffs. Radiation burden from modern chest CT is modest but not negligible, particularly if serial imaging becomes protocolized. Cost and access may also limit feasibility in some health systems. Whether QCT-guided monitoring improves patient-centered outcomes prospectively remains unknown.
Finally, although the study identifies thresholds that are clinically and prognostically relevant, thresholds should not be treated as absolute biological boundaries. Disease progression in IPF exists on a continuum, and clinical decisions still require synthesis of symptoms, physiology, imaging, oxygen needs, and comorbidities.
How This Fits With Current IPF Care
Current international guidance for IPF emphasizes multidisciplinary diagnosis, antifibrotic therapy for appropriate patients, longitudinal monitoring, and timely transplant consideration. Physiologic progression, especially FVC decline, remains a major marker of worsening disease. The present study adds evidence that CT-derived fibrosis progression can function as another dynamic marker with real prognostic content.
This is particularly relevant as radiology and respiratory medicine increasingly incorporate machine learning and automated image analysis into practice. In IPF, where serial subtle changes may be difficult to quantify visually, algorithmic fibrosis measurement may offer a more reproducible way to track disease burden. The study therefore sits at the intersection of respiratory medicine and quantitative imaging, with potentially broader implications for other fibrosing interstitial lung diseases if similar validation work is performed.
Practical Takeaways for Clinicians
For pulmonologists, the main message is that a 1-year increase in QCT-derived fibrosis score is not just an imaging curiosity; it is associated with clinically important outcomes. A rise approaching 4% appears especially informative for adverse prognosis.
For thoracic radiologists, the study reinforces the value of structured longitudinal comparison and supports development of standardized quantitative workflows. Reporting systems may eventually include both absolute fibrosis extent and annual percentage change, analogous to interval volumetric reporting in oncology imaging.
For multidisciplinary interstitial lung disease teams, these findings suggest a potential decision-support role for QCT in patients with discordant clinical data, uncertain progression, or borderline timing for transplant referral. However, the data do not yet justify making major treatment decisions on QCT alone without corroborating clinical context.
Future Directions
The most important next step is prospective validation. Studies should test whether standardized annual QCT monitoring improves decision-making, reduces uncertainty, or enriches trial endpoints beyond conventional lung function measures. Head-to-head comparisons across software platforms will also be critical, because threshold portability is essential for broad adoption.
Another promising avenue is multimodal prediction. Combining ΔFS with serial FVC, DLco, symptom burden, home spirometry, oxygen requirement, and blood biomarkers may yield superior risk models compared with any single measure alone. QCT may also prove useful for identifying treatment responders or for serving as an imaging endpoint in trials of antifibrotic combinations and novel anti-fibrotic or regenerative therapies.
Conclusion
Kim and colleagues provide clinically relevant evidence that 1-year progression in quantitative CT-derived fibrosis extent carries meaningful prognostic information in idiopathic pulmonary fibrosis. Anchor-based MCID estimates ranged from 2.72% to 4.52%, while a prognostic threshold of ΔFS of at least 4.05% most consistently identified patients at higher risk of death or transplantation. The findings support the growing role of quantitative imaging as a complement to pulmonary function testing in longitudinal IPF care. Before routine implementation, prospective validation, technical standardization, and demonstration of management impact will be important. Even so, this study moves the field closer to a practical answer to a long-standing question: how much CT-measured fibrotic progression is enough to matter.
Funding and Trial Registration
The abstract provided does not report funding information or a ClinicalTrials.gov registration number. The study was retrospective in design.
References
Kim DY, Kim MJ, Oh YJ, Hwang HJ, Chae EJ, Lee SM, Park S, Lee JH, Seo JB, Lee HY, Kim HC, Choe J. Meaningful Thresholds for Change in Quantitative CT-Derived Fibrosis Extent in Idiopathic Pulmonary Fibrosis. American Journal of Respiratory and Critical Care Medicine. 2026 May 24. PMID: 42178804.
Raghu G, Remy-Jardin M, Richeldi L, Thomson CC, Inoue Y, Johkoh T, et al. Idiopathic Pulmonary Fibrosis (An Update) and Progressive Pulmonary Fibrosis in Adults: An Official ATS/ERS/JRS/ALAT Clinical Practice Guideline. American Journal of Respiratory and Critical Care Medicine. 2022;205(9):e18-e47.
Richeldi L, du Bois RM, Raghu G, Azuma A, Brown KK, Costabel U, et al. Efficacy and Safety of Nintedanib in Idiopathic Pulmonary Fibrosis. New England Journal of Medicine. 2014;370(22):2071-2082.
King TE Jr, Bradford WZ, Castro-Bernardini S, Fagan EA, Glaspole I, Glassberg MK, et al. A Phase 3 Trial of Pirfenidone in Patients with Idiopathic Pulmonary Fibrosis. New England Journal of Medicine. 2014;370(22):2083-2092.
du Bois RM, Weycker D, Albera C, Bradford WZ, Costabel U, Kartashov A, et al. Forced Vital Capacity in Patients with Idiopathic Pulmonary Fibrosis: Test Properties and Minimal Clinically Important Difference. American Journal of Respiratory and Critical Care Medicine. 2011;184(12):1382-1389.