99mTc-FAPI SPECT/CT for predicting disease progression in fibrosing interstitial lung diseases: study protocol of a single-centre prospective cohort study

Abstract

Introduction

Fibrosing interstitial lung diseases (F-ILD) affect substantial patient populations with significant mortality and socioeconomic burden. Approximately one-third of patients with F-ILD develop a progressive fibrosing phenotype with poor outcomes. Current monitoring methods, including pulmonary function testing and high-resolution CT (HRCT), often lag behind increases in fibrotic activity, limiting early risk identification and optimal treatment timing. Fibroblast activation protein (FAP) imaging can capture fibrotic activity at the pathological level. This study aims to evaluate the incremental value of ⁹⁹ᵐTc-FAPI single-photon emission CT (SPECT)/CT for early warning of disease progression in F-ILD and develop a predictive model integrating fibroblast activation protein inhibitor (FAPI) quantitative parameters with clinical variables.

Methods and analysis

This prospective, single-centre, observational cohort study will recruit 380 adult patients with confirmed F-ILD diagnoses from December 2024 to June 2027, with 12-month follow-up. Comprehensive baseline evaluation includes pulmonary function testing, 6-min walk test, laboratory investigations and ⁹⁹ᵐTc-FAPI SPECT/CT imaging. Follow-up visits at 3, 6 and 12 months will assess clinical outcomes. The primary outcome is disease progression within 12 months, defined as ≥5% forced vital capacity decline, ≥10% diffusing capacity of the lungs for carbon monoxide decline, increased fibrosis on HRCT, acute exacerbation or death. Cox proportional hazards regression models will be used for predictive model construction, with internal validation using bootstrap resampling and temporal split methods.

Ethics and dissemination

The study protocol has been approved by the Ethics Committee of Fujian Provincial Hospital (K2024-11-010) and will be conducted according to the Declaration of Helsinki and Good Clinical Practice guidelines. Written informed consent will be obtained from all participants. Results will be disseminated through peer-reviewed publications and academic conference presentations. De-identified individual participant data will be available upon reasonable request after primary publication.

Trial registration number

Chinese Clinical Trial Registry ChiCTR2400093808.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Prospective cohort design enabling systematic validation of fibroblast activation protein imaging associations with clinical outcomes.

• Broad inclusion of multiple fibrosing interstitial lung disease subtypes with comprehensive baseline and longitudinal assessment.

• Novel application of quantitative ⁹⁹ᵐTc-FAPI single-photon emission CT (SPECT)/CT methodology offering practical advantages over positron emission tomography/CT for clinical implementation.

• Inherent spatial resolution limitations of SPECT/CT imaging potentially affecting sensitivity for small fibrotic lesions.

• Single-centre design without external validation cohort requiring future confirmation in multicentre studies.

Introduction

Fibrosing interstitial lung diseases (F-ILD) represent a heterogeneous group of disorders characterised by pulmonary interstitial inflammation and progressive fibrosis, encompassing subtypes such as idiopathic pulmonary fibrosis (IPF) and connective tissue disease-associated interstitial lung disease (CTD-ILD) and chronic hypersensitivity pneumonitis. The pathological hallmark involves dysregulated injury-repair mechanisms leading to persistent activation of myofibroblasts, which drive extracellular matrix deposition and progressive destruction of the lung architecture, resulting in irreversible ventilatory and diffusion impairments and an increased risk of early mortality.^{1 2} Clinical and real-world data demonstrate that approximately one-third of patients with F-ILD develop a progressive fibrosing phenotype, exhibiting disease behaviour similar to IPF and associated with poor survival outcomes. In a UK multicentre study, approximately 15% of newly referred non-IPF fibrosing ILD patients developed a progressive phenotype during follow-up.^{3 4} Progressive fibrosing interstitial lung disease (PF-ILD) imposes not only significant mortality and healthcare burden but also substantial socioeconomic costs: within the European Economic Area, the estimated patient populations for non-IPF progressive pulmonary fibrosis (PPF) and systemic sclerosis-associated ILD (SSc-ILD) are approximately 86 794 and 13 221, respectively, corresponding to indirect costs of approximately €1.433 billion and €0.220 billion.⁵

Current standard treatment primarily relies on antifibrotic agents, with nintedanib and pirfenidone demonstrated to slow forced vital capacity (FVC) decline and disease progression in IPF and selected PF-ILD populations. Nintedanib has also received approval for SSc-ILD and PF-ILD.^6–8 However, conventional monitoring predominantly depends on pulmonary function testing (such as FVC decline of 5% over 12 months as a predictor of adverse outcomes) and high-resolution CT (HRCT) structural assessment, which often lag behind increases in fibrotic ‘activity’ and have limitations for early risk identification and determining the optimal timing for treatment.^{1 9}

Fibroblast activation protein (FAP) serves as a key marker of activated fibroblasts, and FAP inhibitor (FAPI) molecular imaging can capture fibrotic activity at the pathological level with quantitative expression, thereby providing information that complements functional and structural assessments.^10–12 Previous studies suggest that 68Ga-FAPI PET/CT can identify fibrotic activity in SSc-ILD and animal/human fibrosis models with correlations to functional or risk assessments.^13–15 Concurrently, ^99mTc-FAPI single-photon emission CT (SPECT/CT) offers advantages of higher equipment accessibility, lower cost and easier implementation. Preliminary clinical studies have observed heterogeneous yet quantifiable pulmonary parenchymal uptake in IPF with parametric analysis capabilities which establishes a foundation for clinical translation.^{16 17} To ensure research design and endpoint selection align with international guidelines and enhance methodological rigour of predictive modelling, this study references the American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Asociación Latinoamericana de Tórax clinical practice guidelines for IPF and PPF, as well as standardised frameworks for clinical prediction model development and validation in establishing outcomes, thresholds and validation procedures.^{18 19} However, current evidence primarily consists of small-sample, single-centre or cross-sectional/retrospective studies. To date, FAPI imaging studies have largely examined individual ILD subtypes in isolation, most notably IPF and CTD-ILD, and prospective data spanning the full spectrum of fibrosing ILD remain scarce. Furthermore, no multivariable prediction model integrating FAPI-derived quantitative parameters has been developed or externally validated.^{13–17 20}

Based on these considerations, we propose to conduct a prospective study to systematically evaluate the incremental value of ^99mTc-FAPI SPECT/CT for early warning of disease progression in F-ILD. We aim to construct a predictive model integrating FAPI quantitative parameters with clinical and functional variables, and validate early warning thresholds within a 12-month composite outcome framework, with the goal of providing evidence-based support for precision management transitioning from ‘passive monitoring’ to ‘proactive warning and intervention’.^{18 19} The enrolment of a heterogeneous F-ILD cohort spanning multiple subtypes was guided by the need to improve the generalisability of the prediction model and to permit prespecified subgroup analyses by ILD subtype.

Methods and analysis

Study design

This is a prospective, single-centre, observational cohort study designed to develop and internally validate a 12-month disease progression prediction model for patients with F-ILD based on ^99mTc-FAPI SPECT/CT imaging. The study will be conducted at Fujian Provincial Hospital, a tertiary teaching hospital with extensive experience in the diagnosis and treatment of interstitial lung diseases, located in Fuzhou, Fujian Province, China. Patient recruitment is planned from December 2024 to June 2027, with each participant followed for a minimum of 12 months. The entire study is expected to conclude by June 2028. All study procedures will be conducted in accordance with the Declaration of Helsinki and Good Clinical Practice (GCP) guidelines. All participants will provide written informed consent prior to enrolment.

Study population

Eligible participants are adults aged 18 years or older with a confirmed diagnosis of F-ILD, established through multidisciplinary evaluation incorporating clinical, physiological and HRCT criteria in accordance with current international guidelines. Qualifying diagnoses include IPF, fibrosing nonspecific interstitial pneumonia, chronic hypersensitivity pneumonitis, CTD-ILD and unclassifiable F-ILD.^{21 22} Diagnosis must have been made within the 6 months preceding enrolment. Critically, participants must not fulfil criteria for PPF at the time of enrolment. All participants must demonstrate sufficient cognitive capacity to understand the study procedures and provide written informed consent (consent form provided as an online supplemental file).

Exclusion criteria are as follows: fulfilment of PPF diagnostic criteria at screening; acute exacerbation of interstitial lung disease at screening; acute-phase hypersensitivity pneumonitis, drug-induced lung injury or active pulmonary infection; concurrent enrolment in another interventional clinical trial; planned lung transplantation within 6 months of screening; life expectancy below 6 months attributable to non-pulmonary comorbidities; known hypersensitivity to ⁹⁹mTc-FAPI or any of its constituents; severe renal impairment (estimated glomerular filtration rate<30 mL/min/1.73 m²); inability to complete protocol-required assessments, including pulmonary function testing and the 6-min walk test; and pregnancy or lactation.

Study procedures

All enrolled patients will undergo comprehensive baseline evaluation, including detailed medical history collection encompassing occupational exposure history, smoking history, family history and comorbidities. Following physical examination, complete pulmonary function testing will be performed including FVC, carbon monoxide diffusion capacity, total lung capacity and residual volume to total lung capacity ratio. Six-minute walk test and arterial blood gas analysis will be conducted according to standardised protocols. Laboratory investigations will include complete blood count, hepatic and renal function tests, connective tissue disease screening, and fibrosis serum biomarker Krebs von den Lungen-6 (KL-6). Cardiovascular assessment will include electrocardiography and echocardiography. Patient-reported outcome measures will be assessed using the modified Medical Research Council (mMRC) dyspnoea scale and the Living with Pulmonary Fibrosis questionnaire. Blood samples will be collected and stored in a biobank for future research. Baseline imaging assessment will include ^99mTc-FAPI SPECT/CT and HRCT.

The ^99mTc-FAPI SPECT/CT imaging protocol will use a novel OncoFAP targeting molecule to prepare hydrazinonicotinamide (HYNIC)-oncoFAP (molecular formula: C36H34F2N10O8S; molecular weight: 804.79 Da (HOFAP)) under Good Manufacturing Practice (GMP) standards, prepared using Trinice/triphenylphosphine-3,3’,3’'-trisulfonic acid trisodium salt labelling technology under GMP conditions to create HOFAP kits. HOFAP kits are provided by the Medical Isotope Research Center of Peking University. Imaging agent injection solutions are prepared at the SPECT centre of the Nuclear Medicine Department of Fujian Provincial Hospital. Imaging is performed using the GE Discovery NM/CT 670Pro SPECT/CT system from the USA. After intravenous injection of ^99mTc-FAPI (740 MBq), patients are instructed to drink plenty of water, followed by chest SPECT/CT imaging 2 hours post-injection, with bladder emptying performed before imaging. Images are acquired using low-energy high-resolution collimators with acquisition parameters: tomographic SPECT/CT imaging matrix of 128×128, magnification factor of 1.0, detector rotation of 360° (30 s/frame, 60 frames total); co-registered CT scanning performed at 130 kV and 60 mA. Quality control of acquired images is performed by a nuclear medicine technologist with over 5 years of experience, evaluating patient movement, metallic interference and excessive attenuation correction. Quality-controlled images are interpreted in two groups, with each group comprising two nuclear medicine physicians with over 5 years of experience who independently review images with knowledge of clinical information, with discussions to reach a consensus in case of disagreement.

Follow-up assessments are scheduled at 3, 6, 9 and 12 months post-baseline, with time windows of ±2 weeks for the first three assessments and ±1 month for the final assessment (table 1). Each assessment comprises clinical evaluation, pulmonary function testing, 6-min walk testing, treatment adherence review, adverse event monitoring and serum biomarker measurement. HRCT is performed at 6 and 12 months, and ^99mTc-FAPI SPECT/CT imaging is conducted at 12 months. Unscheduled assessments will be arranged should clinical deterioration occur. All participants will continue to receive standard care from the hospital respiratory medicine department after study completion.

Table 1. Parameters for data collection and follow-up plan.

Assessment/visit	Visit 0	Visit 1	Visit 2	Visit 3	Visit 4
Study timeline	Baseline	Follow-up 1	Follow-up 2	Follow-up 3	Follow-up 4
Month	0	3 (±2 weeks)	6 (±2 weeks)	9 (±2 weeks)	12 (±1 months)
Informed consent	X
Inclusion/exclusion criteria	X
Demographics	X
Disease classification and diagnosis	X
Comorbidities	X	X	X	X	X
Treatment information	X	X	X	X	X
Clinical symptom assessment
mMRC Dyspnoea scale	X	X	X		X
Pulmonary Fibrosis Quality of Life Questionnaire	X				X
Pulmonary function tests
FVC, DLCO, TLC	X	X	X		X
GAP score	X	X	X		X
6-min walk test	X	X	X		X
SpO₂	X	X	X		X
KL-6	X	X	X		X
HRCT scan	X		X		X
99mTc-FAPI SPECT/CT	X				X
Adverse event assessment		X	X	X	X
Disease progression assessment					X
Survival status assessment		X	X	X	X
Blood sample collection	X		X		X

DLCO, diffusion capacity of the lungs for carbon monoxide; FVC, forced vital capacity; GAP, gender, age and physiology; HRCT, high-resolution CT; KL-6, Krebs von den Lungen-6; mMRC, Modified Medical Research Council; 99mTc-FAPI, ⁹⁹ᵐTc-labelled fibroblast activation protein inhibitor; SPECT/CT, single-photon emission CT/CT; SpO₂, peripheral oxygen saturation; TLC, total lung capacity.

Outcome measures

The primary outcome is the occurrence of disease progression within 12 months. Disease progression is determined by defined as meeting any of the following criteria within 12 months: (1) Absolute decline in FVC≥5% of predicted value or absolute decline in diffusing capacity of the lungs for carbon monoxide (DLCO)≥10% of predicted value; (2) increased extent of fibrosis on HRCT; (3) an acute exacerbation or (4) death. Two independent clinicians will assess disease progression, with adjudication by a third senior expert in cases of disagreement.

Secondary outcomes include annual decline rates of FVC and diffusion capacity, time to first acute exacerbation, overall survival and correlations between FAPI uptake parameters and clinical outcomes. Exploratory outcomes include correlations between ^99mTc-FAPI uptake patterns and serum fibrosis biomarkers, development of comprehensive prediction models combining imaging features with clinical and laboratory parameters, F-ILD phenotype identification based on FAPI uptake patterns, radiation dose assessment and safety analysis of ^99mTc-FAPI SPECT/CT and characterisation of population-specific ^99mTc-FAPI SPECT/CT manifestations in Chinese patients with F-ILD.

Sample size calculation

Sample size was determined according to the events-per-variable criterion, which requires at least 10 outcome events per candidate predictor for stable multivariable prediction model development. Eight candidate predictors were prespecified: SUVmean, total lesion FAPI uptake, FVC% predicted, DLCO% predicted, age, ILD subtype, Usual Interstitial Pneumonia-like pattern on HRCT and extent of fibrosis on HRCT. This yielded a minimum requirement of 80 progression events.

Published data indicate that approximately 18–32% of patients with non-IPF F-ILD develop PPF despite treatment.²² A 12-month progression rate of 25% was adopted as a conservative estimate. With an anticipated dropout and loss-to-follow-up rate of 15%, at least 377 participants would be needed to accumulate 80 progression events. The target sample size was therefore set at 380, which provides at least 80% statistical power at a two-sided significance level of 0.05. All calculations were performed using R software V.4.2.0 with relevant statistical packages.

Data collection and management

Data will be collected using an electronic data capture (EDC) system with built-in logic checks to ensure data quality. Each patient will be assigned a unique identifier to maintain confidentiality. A dedicated clinical research coordinator will be responsible for data entry and quality control. An independent contract research organisation will conduct periodic monitoring to verify source data and ensure protocol compliance.

All study materials will be securely stored at the research centre, with electronic data protected by password access and regularly backed up. No formal interim analyses are planned due to the observational nature and 12-month follow-up period of this study. Biological samples will be preserved in a biobank for future research with appropriate consent.

Statistical analysis

Descriptive statistics will be used to summarise demographic and clinical characteristics, with continuous variables presented as mean±SD or median with IQR according to data distribution, and categorical variables presented as frequencies and percentages. Predictive model construction will use Cox proportional hazards regression models with stepwise regression for variable selection. The first 70% of enrolled patients will constitute the training cohort, and the remaining 30% will serve as the internal validation cohort.

Model performance evaluation will include assessment of discrimination using Harrell’s C-index, calibration using calibration plots and Hosmer–Lemeshow test and clinical utility using decision curve analysis. Internal validation will employ bootstrap resampling with 1000 iterations and temporal split validation methods. The primary analysis will be based on complete cases, with sensitivity analysis using multiple imputation to handle missing data. Subgroup analyses will be conducted according to F-ILD subtypes. All statistical analyses will be performed using R software V.4.2.0, with statistical significance set at p<0.05.

Withdrawal and discontinuation

Participation is entirely voluntary, and all participants may withdraw at any time without providing reasons and without prejudice to their subsequent clinical care. Upon withdrawal, participants will be offered two options: (1) consent to the retention and use of data collected up to the point of withdrawal, which will be included in the final analysis; or (2) request that all previously collected data be excluded from analysis and handled in accordance with applicable data protection regulations. Both options will be explained during the informed consent process and documented in the case report form. The withdrawal rate will be reported in the final study publication.

Ethics and dissemination

The study will be conducted in strict accordance with the Declaration of Helsinki and relevant Chinese regulations. The study protocol has been approved by the Ethics Committee of Fujian Provincial Hospital (approval No.K2024-11-010). Any protocol amendments will be submitted to the ethics committee of Fujian Provincial Hospital and trial registry, and communicated to all study participants as required. Study results will be disseminated through publication in peer-reviewed journals and presentation at academic conferences. De-identified individual participant data will be made available upon reasonable request after primary publication.

Patient and public involvement

Patients and the public were not involved in the design of the study, nor were they involved in the recruitment and conduct of the study.

Discussion

The precision management of F-ILD has long been constrained by several challenges, including difficulties in early diagnosis, an insufficient quantification of disease activity and the lag of functional outcomes, resulting in a lack of sensitive evidence for selecting treatment strategies and their timing.^{1 9 20} FAP-targeted molecular imaging characterises the spatial distribution and intensity of activated fibroblasts, reflecting ‘fibrotic activity’ at the biological level and providing complementary information to FVC/HRCT assessments. Multiple preclinical and early clinical studies have demonstrated that FAPI uptake correlates with disease activity or future functional decline, indicating its potential as an imaging biomarker for risk stratification and therapeutic monitoring.^10–15 Among FAPI-labelled tracers, ⁹⁹ᵐTc-labelled compounds have attracted particular interest in fibrosing ILD. Wang et al demonstrated that ⁹⁹ᵐTc-HFAPI uptake correlated significantly with pulmonary function indices and HRCT fibrosis scores in CTD-ILD, and these associations were consistently stronger than those observed for ¹⁸F-FDG, suggesting that FAP-targeted imaging may be better suited than metabolic imaging to the characterisation of fibrotic rather than inflammatory disease activity.²³ A pilot study subsequently confirmed that quantifiable parenchymal uptake of ⁹⁹ᵐTc-labelled FAPI is achievable in F-ILD, with parametric analysis demonstrating the technical feasibility of this approach in clinical settings.²⁴ ⁹⁹ᵐTc-FAPI SPECT/CT also carries practical advantages over ⁶⁸Ga- or ¹⁸F-labelled PET/CT, namely wider equipment availability and lower cost, which have direct bearing on the feasibility of serial imaging for disease monitoring. Technical validation of ⁹⁹ᵐTc-labelled FAPI tracers as a clinically viable alternative to PET has been provided in oncological settings: a prospective head-to-head comparison of [⁹⁹ᵐTc]Tc-DP-FAPI SPECT/CT and [⁶⁸Ga]Ga-FAPI-04 PET/CT in patients with gastrointestinal tumours demonstrated equivalent sensitivity for primary lesion detection, though SPECT showed reduced performance in lymph node characterisation.²⁵ Whether analogous trade-offs apply in fibrosing ILD, where the imaging target is diffuse bilateral interstitial disease rather than discrete nodal deposits, remains to be established. Critically, none of the existing studies have examined whether baseline ⁹⁹ᵐTc-FAPI SPECT quantitative parameters can predict subsequent disease progression, and no multivariable prediction model incorporating FAPI-derived indices has been developed or validated in this population.^{16 17}

The present study enrols a multi-subtype F-ILD cohort and evaluates the relationship between baseline ⁹⁹ᵐTc-FAPI quantitative indices and 12-month outcomes defined according to current international guideline criteria for PPF. The primary analytical goal is to develop and internally validate composite prediction models capable of supporting individualised risk stratification and treatment timing decisions. The study adheres to established methodological frameworks for clinical prediction model development, with the intention of strengthening reproducibility and providing a foundation for future external validation and direct comparison with PET-based FAPI imaging.^18–20 Limitations that must be objectively acknowledged include the spatial resolution constraints of SPECT/CT relative to PET/CT, which may affect sensitivity for small or marginal lesions; the impact of the single-centre design and sample size on external validity; the inclusion of a heterogeneous fibrosing ILD cohort, which, while scientifically justified by the shared progressive fibrosing phenotype, limits subtype-specific conclusions, and external validation in larger, subtype-specific cohorts including IPF or PPF will be necessary to confirm the generalisability of the predictive model; the absence of systematic histopathological data precludes formal correlation between SPECT/CT-derived quantitative parameters and tissue-level FAP expression, which warrants dedicated investigation in future biopsy-confirmed cohorts; and the need for deeper integration of imaging parameters with serum markers, breathomics and multi-omics data to enhance model robustness and generalisability.^{6–9 16 17 19 20} Future directions include conducting multicentre prospective validation to confirm reproducibility of imaging thresholds, combining multi-omics and artificial intelligence approaches to improve predictive performance, and validating the incremental value of FAPI parameters for explaining early responses to therapeutic changes and predicting long-term outcomes in dynamic monitoring applications during antifibrotic or immunosuppressive therapy.^18–20

Overall, the quantification capabilities of ^99mTc-FAPI SPECT/CT centred on ‘activity’ through FAP biology have the potential to address blind spots in traditional F-ILD assessment, driving a clinical decision-making transition from ‘passive monitoring’ to ‘proactive warning and precision intervention’.