Upfront combination therapy with nintedanib and anti-inflammatory agents for progressive pulmonary fibrosis: a multicentre, single-arm phase 2 study (TOP-ILD)

Abstract

Objective

Progressive pulmonary fibrosis (PPF) is a chronic interstitial lung disease (ILD) characterised by fibrotic progression and poor prognosis, with effective treatment strategies for previously untreated patients remaining unclear. This study evaluated the efficacy and safety of upfront combination therapy with anti-inflammatory and antifibrotic agents in previously untreated PPF patients.

Methods

This multicentre, single-arm phase 2 study enrolled 34 patients with ILD (including unclassifiable idiopathic interstitial pneumonia, idiopathic nonspecific interstitial pneumonia, fibrotic hypersensitivity pneumonitis and rheumatoid arthritis-associated ILD) all with evidence of PPF. Tacrolimus (0.0375 mg·kg⁻¹ twice daily) and prednisolone (10 mg once daily) were initiated on day 1, with nintedanib (150 mg twice daily) added on day 8. The tacrolimus dosage was adjusted to maintain blood trough levels. The primary end-point was the change in the relative decline slope for forced vital capacity % predicted (%FVC) between before and after treatment.

Results

The protocol treatment was associated with a substantial improvement in the relative %FVC decline slope, from −20.9% per year before to +11.2% per year after treatment. Subgroup analysis revealed greater improvement in patients with an increased lymphocyte percentage in bronchoalveolar lavage fluid or elevated blood biomarkers. Adverse events, such as diarrhoea (67.6%) and hepatic dysfunction (29.4%), were manageable, with no severe cases or treatment discontinuations.

Conclusion

Early combination therapy with tacrolimus, prednisolone and nintedanib was associated with improved pulmonary function and was well tolerated in previously untreated PPF patients. Our findings suggest the potential of this regimen as an initial treatment strategy, but further validation in larger randomised controlled trials is warranted.

Shareable abstract

Early combination therapy with tacrolimus, prednisolone and nintedanib significantly improves pulmonary function in previously untreated patients with progressive pulmonary fibrosis. The regimen was well tolerated, with manageable adverse effects. https://bit.ly/46SCH4r

Introduction

Progressive pulmonary fibrosis (PPF) is characterised by progressive fibrotic changes in individuals with interstitial lung disease (ILD) of known or unknown etiology, excluding idiopathic pulmonary fibrosis (IPF). This progression is evidenced by symptomatic, physiological or radiological deterioration [1]. The initial management of PPF focuses primarily on standard treatment targeting the underlying disease. In cases such as unclassifiable idiopathic interstitial pneumonia (IIP), idiopathic nonspecific interstitial pneumonia (iNSIP), fibrotic hypersensitivity pneumonitis (fHP) or rheumatoid arthritis-associated ILD (RA-ILD), immunosuppressive therapies are commonly administered to stabilise or ameliorate the primary disease [2–4]. If corticosteroid efficacy is insufficient or if tapering is required, additional immunosuppressive agents are often introduced [5]. However, evidence-based guidelines for the standardised management of many ILD subtypes remain inadequate.

Nintedanib, an antifibrotic agent, targets key pathogenic pathways related to progressive fibrosis [6]. Given the shared pathophysiological mechanisms and clinical progression between PPF and IPF, nintedanib has shown efficacy and safety for PPF as it has for IPF [7–12]. Current guidelines recommend nintedanib as a treatment option for individuals with PPF who do not respond adequately to standard therapies for fibrotic ILD [1]. However, in the real-world clinical setting, antifibrotic therapy, including nintedanib, is often initiated in individuals for whom impairment of lung function is already advanced. An international phase 3 trial of nintedanib for progressive fibrosing ILD (PF-ILD) showed that it reduced the rate of forced vital capacity (FVC) decline across subgroups stratified by FVC % predicted (%FVC) at baseline [13]. Of note, the inhibitory effect on FVC decline was numerically greater in patients with a higher baseline %FVC, underscoring the potential benefit of initiating antifibrotic therapy at early stages of disease progression [14].

Given this background, we hypothesised that upfront combination therapy with antifibrotic and immunosuppressive agents might provide a superior outcome for management of PPF characterised by overlapping fibrotic and inflammatory pathophysiological features, such as unclassifiable IIP, iNSIP, fHP and RA-ILD. The present study was therefore performed to evaluate the efficacy and safety of concurrent treatment with tacrolimus, prednisolone and nintedanib in an attempt to identify optimal therapeutic strategies for previously untreated PPF.

Methods

Study design and participants

This study (TOP-ILD) was designed as a multicentre, single-arm, nonrandomised phase 2 trial. Individuals aged 20 years or older with a diagnosis of unclassifiable IIP, iNSIP, fHP or RA-ILD based on multidisciplinary discussion (MDD) as well as physician-diagnosed PPF were eligible for the study. Participants were required to meet the following criteria for ILD progression within the previous year: 1) physiological evidence of disease progression (a relative decline in FVC of >5% predicted or a relative decline in diffusing capacity of the lung for carbon monoxide (D_LCO), corrected for haemoglobin, of >15% predicted) and worsening respiratory symptoms; or 2) physiological and radiological evidence of disease progression [1]. Enrolled patients had no prior treatment for ILD and showed features of fibrosing lung disease on high-resolution computed tomography (HRCT), defined as a reticular abnormality with traction bronchiectasis, with or without honeycombing, affecting >10% of the lung. At enrolment, participants were also required to have an FVC of ≥50% of the predicted value and a D_LCO (corrected for haemoglobin) of 30% to 80% of the predicted value.

Exclusion criteria included prior treatment with corticosteroids, immunosuppressive agents (such as tacrolimus, cyclosporine, cyclophosphamide, azathioprine and mycophenolate mofetil) or antifibrotic agents for ILD; a history of acute exacerbation of ILD; predominant airflow obstruction; active malignancy requiring chemotherapy; active infection requiring systemic treatment; other serious medical conditions (such as uncontrolled diabetes, unstable angina, severe pulmonary hypertension, venous thromboembolism, and acute or chronic pancreatitis); risk of bleeding; psychological disorders that might impair study participation; and pregnancy, postpartum status or breastfeeding.

Procedures

Tacrolimus (0.0375 mg·kg⁻¹ twice daily) and prednisolone (10 mg once daily) were initiated on day 1. In accordance with previous reports and the prescribing information in Japan, tacrolimus was initiated at a dosage of 0.0375 mg·kg⁻¹ twice daily [15, 16]. The tacrolimus dosage was adjusted to maintain trough levels in whole blood of 5 to 10 ng·mL⁻¹. Nintedanib (150 mg twice daily) was introduced on day 8. Participants experiencing adverse events were allowed to interrupt treatment or to reduce the nintedanib dose to 100 mg twice daily. Dose reductions or discontinuation of tacrolimus and prednisolone were allowed as deemed appropriate.

The study complied with the Declaration of Helsinki, was approved by the Kyushu University Certified Institutional Review Board for Clinical Trials, and was registered in the Japan Registry of Clinical Trials (jRCTs071230004). All data were monitored centrally by Clinical Research Support Center Kyushu (CReS Kyushu). Written informed consent was obtained from all participants.

Baseline spirometry and chest HRCT were performed before treatment initiation. Spirometry was also conducted at 0, 12 and 24 weeks and chest HRCT at 0 and 24 weeks relative to treatment onset. Data regarding treatment, physical examinations and clinical assessments were collected at each visit. Acute exacerbation of ILD or death was assessed throughout the observation period. Usual interstitial pneumonia (UIP)-like fibrotic patterns on HRCT were defined according to previous trial criteria [8] and on the basis of one of the following: 1) definite honeycombing with basal and peripheral predominance; 2) reticular abnormalities with traction bronchiectasis indicative of fibrosis with a similar distribution; or 3) the absence of atypical features such as nodules or consolidation, with any ground-glass opacities being less extensive than the reticular pattern. Acute exacerbation of ILD was defined according to the following criteria, all of which had to be met within 1 month after exclusion of alternative diagnoses (such as pulmonary infection, pneumothorax, malignancy, pulmonary embolism and heart failure): 1) worsening dyspnoea; 2) new bilateral ground-glass opacities or consolidation superimposed on preexisting reticular or honeycomb patterns; and 3) a decrease in arterial oxygen tension (P_aO2) levels of >10 mmHg [17]. Safety assessments included clinical and laboratory evaluations, with adverse events coded according to the Medical Dictionary for Regulatory Activities, version 27.1. Data were collected by a web-based electronic data capture system, Viedoc4 (Viedoc Technologies AB, Uppsala, Sweden).

Outcomes

The primary end-point of the study was the change in the relative %FVC decline slope between before and after treatment (supplementary figure S1): the pretreatment relative %FVC decline slope was calculated as: ((%FVC at 0 weeks – %FVC at baseline)/%FVC at baseline)×365.25/observation period (days). The post-treatment relative %FVC decline slope was calculated as: ((%FVC at 24 weeks – %FVC at 0 weeks)/%FVC at 0 weeks))×365.25/observation period (days). The anticipated change in the relative %FVC decline slope was estimated to be 3.5% on the basis of an expected treatment response similar to that observed in an international phase 3 study of nintedanib in patients with progressive fibrosing ILD (PF-ILD) [8]. With a standard deviation of 7.6% derived from this previous nintedanib trial, we established a half-width of 2.7% for the 95% confidence interval (CI) of the change in relative %FVC decline slope. Given these assumptions, the study was planned with a set sample size of 32 patients. Key secondary end-points included the absolute change in %FVC and % D_LCO from baseline to 24 weeks, safety and toxicity, and time to the first acute exacerbation of ILD or death.

Statistical analysis

Point estimates and their 95% CIs based on the t-distribution were calculated for the changes in relative %FVC decline slope, %FVC and % D_LCO. If the 24-week measurement was not available, the most recent measurement and observation time up to that point were used to calculate the relative %FVC decline slope. Time to first acute exacerbation of ILD or death was estimated with the Kaplan–Meier method. Subgroup analysis was performed for the change in the relative %FVC decline slope and for that in %FVC according to baseline characteristics. All statistical analysis was conducted with SAS version 9.4 (SAS Institute) and R version 4.4.1 (R Foundation).

Results

Patient information

Between April 2023 and January 2024, a total of 34 patients was enrolled across 16 institutions. All 34 patients initiated protocol treatment (figure 1), and their characteristics are summarised in table 1. All participants were diagnosed with ILD by MDD at each institution in accordance with international guidelines [1, 2, 18, 19]. The most common diagnosis was unclassifiable IIP (47.1%), followed by fHP (44.1%), iNSIP (5.9%) and RA-ILD (2.9%). The median patient age was 71 years, 22 patients (64.7%) were male, and 21 (61.8%) had a history of smoking. For the overall study population, %FVC (mean±sd) was 75.2±17.1% and % D_LCO (mean±sd) was 58.2±12.6%. The median percentage of lymphocytes in bronchoalveolar lavage fluid (BALF) was 9.0%. Eight patients (23.5%) tested positive for autoantibodies. A UIP-like fibrotic pattern on HRCT was identified in 20 patients (58.8%), whereas emphysematous changes were observed in seven patients (20.6%). According to IPF guidelines [9], the most frequent radiological pattern was an alternative diagnosis (44.1%), followed by indeterminate for UIP (32.4%), probable UIP (14.7%) and UIP (8.8%) (supplementary table S1). According to hypersensitivity pneumonitis (HP) guidelines [18], radiological patterns were classified as typical HP (11.8%), compatible with HP (29.4%), and indeterminate for HP (58.8%). Histological evaluations, including cryobiopsy or surgical biopsy, were conducted in 23 patients. The most frequent histological pattern, according to the IPF guidelines [1], was indeterminate for UIP (47.8%), followed by an alternative diagnosis (34.8%).

FIGURE 1.

Patient flow diagram for the study.

TABLE 1.

Baseline characteristics of the study population (n=34)

Age years, median (range)	71 (40–83)
Sex, n (%)
Male	22 (64.7)
Female	12 (35.3)
Clinical ILD diagnoses, n (%)
Unclassifiable idiopathic interstitial pneumonia	16 (47.1)
Fibrotic hypersensitivity pneumonitis	15 (44.1)
Idiopathic nonspecific interstitial pneumonia	2 (5.9)
Rheumatoid arthritis-associated ILD	1 (2.9)
Smoking, n (%)
Current or former smoker	21 (61.8)
Never-smoker	13 (38.2)
Body mass index, kg·m−2, median (range)	24.6 (17.5–33.3)
FVC % predicted, mean±sd	75.2±17.1
DLCO % predicted, corrected for Hb, mean±sd	58.2±12.6
UIP-like fibrotic pattern on HRCT, n (%)	20 (58.8)
Emphysematous change, n (%)	7 (20.6)
BALF lymphocytosis %, mean (range)	9.0 (0.4–78.8)
mMRC scale, 0/1/2/3/4, n	4/20/8/1/1
Serum C-reactive protein, mg·dL−1, mean±sd	0.64±1.39
Autoantibody positive, n (%)	8 (23.5)
Anti-nuclear antibody (≥1:320), n (%)	4 (11.8)
Anti-CCP antibody, n (%)	5 (14.7)
Anti-RNP antibody, n (%)	1 (2.9)
Anti-topoisomerase I antibody, n (%)	1 (2.9)
Anti-MDA5 antibody, n (%)	1 (2.9)
Serum KL-6, U·mL−1, mean±sd	1227±855
Serum SP-D, ng·mL−1, mean±sd	337±217
Serum SP-A, ng·mL−1, mean±sd	80.4±51.6

31 patients underwent bronchoalveolar lavage. Data for KL-6 and SP-D only shown for 32 and 31 patients, respectively. ILD: interstitial lung disease; FVC: forced vital capacity; D_LCO: diffusing capacity of the lung for carbon monoxide; UIP: usual interstitial pneumonia; HRCT: high-resolution computed tomography; BALF: bronchoalveolar lavage fluid; mMRC: modified Medical Research Council; CCP: cyclic citrullinated peptide; RNP: ribonucleoprotein; MDA5: melanoma differentiation-associated protein 5; KL-6: Krebs von den Lungen-6; SP-D: surfactant protein D; SP-A: surfactant protein A.

Treatment delivery

All 34 patients initiated treatment with tacrolimus and prednisolone, and 33 subsequently received nintedanib in combination with tacrolimus and prednisolone (figure 1). One patient was unable to start nintedanib within 28 days after initiation of tacrolimus and prednisolone because of pneumothorax. 32 patients (94.1%) completed the 24-week protocol treatment, with one patient withdrawing consent (figure 1). Treatment delivery is summarised in table 2.

TABLE 2.

Treatment delivery

Time	Medication	n (%)	Reason for omission
Day 1	TAC+PSL	34 (100)
Day 8	NTD+TAC+PSL (Protocol treatment)	33 (97.1)	Pneumothorax
Day 29	NTD+TAC+PSL	33 (97.1)
Day 85	NTD+TAC+PSL	33 (97.1)
Day 169	NTD+TAC+PSL	32 (94.1)	Withdrew

TAC: tacrolimus; PSL: prednisolone; NTD: nintedanib.

The dosage of each drug over 24 weeks is detailed in supplementary table S2. At week 24, nintedanib was administered at 150 mg twice daily in 22 patients and at 100 mg twice daily in 11 patients. The most common reason for nintedanib dose reduction was liver dysfunction, which was observed in seven patients and in most instances occurred within the first 2 months. Other reasons for such dose reduction included diarrhoea (two patients), appetite loss (one patient) and mediastinal emphysema (one patient). Tacrolimus dosage was reduced in three patients, as a result of hepatic or renal dysfunction, whereas prednisolone dosage was reduced in four patients because of elevated intraocular pressure, femoral head necrosis, glucose intolerance or liver dysfunction. No patient discontinued treatment as a result of drug-related side-effects.

Efficacy

The 33 patients who received nintedanib, tacrolimus and prednisolone were included in the efficacy analysis (figure 1). For the primary end-point, the relative %FVC decline slopes before and after treatment were calculated and compared (supplementary figure S1). The change in the relative %FVC decline slope between before and after treatment intervention was 32.1% per year (95% CI 17.2–47.0% per year), which met the primary objective of the study. The relative %FVC decline slope (mean±sd) was −20.9±27.5% per year before treatment and +11.2±20.0% per year after treatment (figure 2a). The absolute changes in %FVC from baseline to weeks 12 and 24 were 3.7% (95% CI 1.4–5.9%) and 3.8% (95% CI 1.1–6.6%), respectively (figure 2b). A forest plot summarising the results of subgroup analysis for the change in the relative %FVC decline slope stratified by factors such as age, clinical diagnosis, UIP-like pattern, radiological pattern based on IPF guidelines, smoking history, body mass index (BMI), %FVC at enrolment, BALF lymphocyte percentage and blood biomarkers (C-reactive protein (CRP), Krebs von den Lungen-6 (KL-6), surfactant protein-D (SP-D), D-dimer and monocyte count) is shown in figure 3. The %FVC decline slope improved after treatment in all subgroups. Patients with BALF lymphocytosis (≥20%) showed a numerically greater improvement in the absolute change in %FVC compared with those with a lower lymphocyte percentage (<20%) (supplementary figures S2 and S3). Greater improvements were also observed in patients with elevated blood biomarkers (CRP, KL-6, SP-D, D-dimer and monocyte count). No acute exacerbations or deaths occurred during the observation period.

FIGURE 2.

a) Change in the relative decline slope of forced vital capacity (FVC) % predicted before and after treatment. b) Absolute change in FVC % predicted during the screening (pretreatment) and treatment periods. Data are presented as mean±sd. PPF: progressive pulmonary fibrosis.

FIGURE 3.

Forest plot of the change in the relative decline slope of forced vital capacity % predicted (%FVC) between before and after treatment as stratified by baseline characteristics. BALF: bronchoalveolar lavage fluid; BMI: body mass index; CRP: C-reactive protein; IIP: idiopathic interstitial pneumonia; ILD: interstitial lung disease; KL-6: Krebs von den Lungen-6; SP-D: surfactant protein-D; UIP: usual interstitial pneumonia.

Safety and tolerability

All 34 patients were included in the safety analysis. Major treatment-emergent adverse events are listed in table 3. The most frequent adverse event was diarrhoea, reported in 23 patients (67.6%), with moderate diarrhoea occurring in four patients (11.8%) and no cases of severe diarrhoea being noted. Hepatic adverse events were also common, with elevations in aspartate aminotransferase and alanine aminotransferase levels being observed in eight patients (23.5%) each and moderate elevations in these enzymes occurring in four (11.8%) and five (14.7%) patients, respectively. Impaired glucose tolerance associated with tacrolimus and prednisolone treatment was noted in 13 patients (38.2%). Two cases of infectious disease, including COVID-19 in one patient, were reported.

TABLE 3.

Most commonly reported treatment-emergent adverse events in all treated patients (n=34)

	Grade, n (%)
Treatment-emergent adverse events	Any grade	Mild	Moderate
Gastrointestinal disorders
Diarrhoea	23 (67.6)	19 (55.9)	4 (11.8)
Constipation	2 (5.9)	2 (5.9)	0 (0.0)
Nausea	2 (5.9)	2 (5.9)	0 (0.0)
Vomiting	2 (5.9)	1 (2.9)	1 (2.9)
Abdominal discomfort	1 (2.9)	1 (2.9)	0 (0.0)
Abdominal distension	1 (2.9)	1 (2.9)	0 (0.0)
Hepatobiliary disorders
Aspartate aminotransferase increased	8 (23.5)	4 (11.8)	4 (11.8)
Alanine aminotransferase increased	8 (23.5)	3 (8.8)	5 (14.7)
Blood bilirubin increased	1 (2.9)	1 (2.9)	0 (0.0)
Hepatic function abnormal	3 (8.8)	2 (5.9)	1 (2.9)
Metabolism and nutrition disorders
Glucose tolerance impaired	13 (38.2)	11 (32.4)	2 (5.9)
Decreased appetite	4 (11.8)	3 (8.8)	1 (2.9)
Infections and infestations
COVID-19	1 (2.9)	1 (2.9)	0 (0.0)
Nasopharyngitis	1 (2.9)	1 (2.9)	0 (0.0)
Renal and urinary disorders
Renal impairment	1 (2.9)	1 (2.9)	0 (0.0)
Ureterolithiasis	1 (2.9)	0 (0.0)	1 (2.9)
Pollakiuria	1 (2.9)	1 (2.9)	0 (0.0)
General disorders and administration site conditions
Fatigue	3 (8.8)	3 (8.8)	0 (0.0)
Fever	3 (8.8)	3 (8.8)	0 (0.0)
Chest discomfort	1 (2.9)	1 (2.9)	0 (0.0)
Investigations
Weight decreased	1 (2.9)	1 (2.9)	0 (0.0)
C-reactive protein increased	2 (5.9)	2 (5.9)	0 (0.0)
Intraocular pressure increased	1 (2.9)	1 (2.9)	0 (0.0)
Respiratory, thoracic and mediastinal disorders
Dyspnoea	2 (5.9)	1 (2.9)	1 (2.9)
Cough	2 (5.9)	2 (5.9)	0 (0.0)
Pneumothorax	2 (5.9)	2 (5.9)	0 (0.0)
Sputum increased	1 (2.9)	1 (2.9)	0 (0.0)
Dysphonia	1 (2.9)	1 (2.9)	0 (0.0)
Nervous system disorders
Taste disorder	1 (2.9)	1 (2.9)	0 (0.0)
Hypoesthesia	1 (2.9)	1 (2.9)	0 (0.0)
Dizziness	1 (2.9)	1 (2.9)	0 (0.0)

Pneumothorax includes pneumothorax and pneumothorax spontaneous. Glucose tolerance impaired includes glucose intolerance, hyperglycaemia, glycosylated haemoglobin increased and steroid diabetes.

Discussion

This study represents the first prospective, multicentre, interventional investigation to evaluate the efficacy and safety of upfront combination therapy with nintedanib, tacrolimus and prednisolone in previously untreated individuals with PPF. This treatment regimen showed substantial efficacy, improving the relative decline slope of %FVC from −20.9% per year to +11.2% per year in a heterogeneous ILD population, including patients with unclassifiable IIP (47.1%) and fHP (44.1%). Although %FVC improved within 12 weeks of treatment onset and was subsequently maintained through 24 weeks, it did not return to preprogression levels. Treatment efficacy was consistent across subgroups stratified by clinical characteristics such as a UIP-like fibrotic pattern on HRCT. Subgroup analysis revealed a greater improvement in %FVC in patients with a higher lymphocyte percentage in BALF or elevated blood biomarkers, suggesting that these characteristics may serve as predictive markers for therapeutic response. The safety and tolerability profiles of the protocol treatment were acceptable, with most adverse events being manageable.

To date, no standardised, prospectively evaluated management strategy exists for PPF, including unclassifiable IIP, iNSIP and fHP. Initial treatment for such patients has primarily relied on anti-inflammatory agents on the basis of retrospective studies. In the case of fHP, retrospective studies have reported an improvement in D_LCO or FVC after 1 year of treatment with mycophenolate mofetil or azathioprine [20–22]. However, evidence supporting a long-term benefit of corticosteroids or their ability to slow fHP progression remains limited [23–25]. Indeed, recent studies have suggested that corticosteroid therapy may be harmful in patients with telomere shortening or specific causative antigens, underscoring the need for prospective verification [26–28]. Unclassifiable IIP and iNSIP share clinical characteristics with collagen vascular disease-associated ILD. In the present study, seven of 18 patients with unclassifiable IIP or iNSIP tested positive for autoantibodies, serological markers of interstitial pneumonia with autoimmune features (IPAF). Anti-inflammatory therapy is often administered as an initial treatment in such cases, but no prospective interventional studies have been conducted. Our study is the first to prospectively demonstrate a substantial improvement in lung function in previously untreated PPF patients receiving a therapeutic regimen including anti-inflammatory agents. A previous retrospective study of IPAF patients with a histological UIP pattern reported a 12% improvement in FVC after 1 year of immunosuppressive therapy [29], consistent with our findings. In our study, comparable treatment responses were observed in patients across all radiological patterns. This finding is consistent with a recent report indicating a lack of correlation between the effectiveness of immunosuppressive agents and the ground-glass opacity to fibrosis ratio on HRCT imaging [30]. Whereas anti-inflammatory treatments are harmful in IPF [31], they may benefit individuals with fibrotic ILD other than IPF when diagnosed confidently by MDD.

In the present study, %FVC showed marked improvement at 3 months after treatment onset and had reached a plateau at a level below the preprogression baseline at 6 months. A similar trend was observed in patients with BALF lymphocytosis, a subgroup expected to derive substantial therapeutic benefit from corticosteroids [24]. These findings suggest that PPF, in particular that associated with underlying diseases such as unclassifiable IIP and fHP, is characterised by an irreversible fibrotic process as well as inflammation. An international phase 3 trial reported a 3.24% annual improvement in relative %FVC decline with nintedanib compared with the control arm in PF-ILD patients previously treated by standard management [8]. A selective phosphodiesterase 4B inhibitor, a potential novel antifibrotic agent, has shown synergistic effects in suppressing the fibrosis process with its anti-inflammatory properties [32]. The favourable outcomes in our study may have resulted from the early initiation of antifibrotic therapy at a time when lung function was preserved as well as potential synergy between anti-inflammatory and antifibrotic agents. The 24-week observation period in the present study provided promising results, and a 52-week follow-up analysis is anticipated to evaluate further the long-term efficacy and safety.

Safety analysis revealed a manageable adverse event profile for the study treatment, consistent with previous findings [15, 33, 34]. Diarrhoea (67.6%) and hepatic dysfunction (29.4%) were the most common adverse events, with moderate diarrhoea occurring in 11.8% of patients. Dose adjustments effectively controlled side-effects. The proportion of patients with increased hepatic enzyme levels in our study was higher than that reported in the pivotal international phase III trials of nintedanib [8]; this may reflect a population-specific phenomenon. Previous reports have suggested a potential predisposition to nintedanib-induced hepatic dysfunction in Japanese individuals, possibly attributable to lower body surface area [35]. Nevertheless, global post-marketing surveillance data for nintedanib indicate an overall rate of hepatic enzyme elevations of 23.3%, comparable to that observed in our study [36], suggesting sufficient tolerability of our early combination therapy despite this potential variance. Of note, no severe adverse events and no treatment discontinuations occurred. In contrast, postmarketing studies of nintedanib in IPF patients revealed a 50% discontinuation rate within 12 months for various reasons [33]. The high treatment continuation rate in our study may have been attributable to factors such as preserved %FVC, concomitant prednisolone use and high BMI [33, 34, 37]. Metabolic complications related to tacrolimus and prednisolone, such as impaired glucose tolerance, were manageable without treatment discontinuation in the present study. The absence of acute exacerbation or mortality during the observation period further supports the tolerability of the study regimen. However, long-term monitoring will be essential to evaluate potential adverse effects, including infections, beyond the 24-week period.

Our study has several limitations. First, it was conducted exclusively in Japan and with a relatively small sample size, which may limit the generalisability of the findings. Second, the absence of a control group complicates attribution of the observed beneficial effects solely to the combination therapy. These results may also be partially influenced by the natural course of the disease or other confounding factors. Although the study adopted a pre-post comparison design, with the preintervention phase for the same patients as a control, this approach has precedent in clinical trials for IPF [38, 39], but it remains a limitation. Third, the 24-week observation period is insufficient for assessment of the long-term efficacy and safety of this combination therapy in ILD, a chronic and progressive disease. Future analyses with extended follow-up periods are planned.

In conclusion, the present study suggests that early initiation of antifibrotic therapy combined with anti-inflammatory therapy may be a potentially effective and tolerable strategy for individuals with PPF characterised by fibrotic patterns. Our findings provide a foundation for future research. Large randomised controlled trials comparing the conventional sequence of anti-inflammatory therapy followed by antifibrotic agents with this upfront combination therapy will be essential to further elucidate the role of this latter regimen in clinical practice.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Figure S1

00697-2025.SUPPLEMENT

Figure S2

00697-2025.SUPPLEMENT2

Figure S3

00697-2025.SUPPLEMENT3

Data availability

Data may be obtained from a third party and are not publicly available.