Outcomes and prognosis of progressive pulmonary fibrosis in patients with antineutrophil cytoplasmic antibody-positive interstitial lung disease

Approximately one-third of fibrosing interstitial lung diseases exhibit progressive pulmonary fibrosis (PPF), a clinicopathological condition distinct yet resembling idiopathic pulmonary fibrosis (IPF). PPF in ANCA-positive ILD (ANCA-ILD) is poorly documented. To clarify incidence, predictors of PPF in ANCA-ILD, and their prognostic impact, 56 patients with ANCA-ILD were followed for ≥ 1 year (April 2004 to April 2021). PPF was defined per ATS/ERS/JRS/ALAT PPF 2022 guideline. We compared PPF and non-PPF in 38 patients with pulmonary function tests and ≥ 1 year follow up. ANCA-ILD (19 male, 19 female; mean age 72 years) comprised 21 patients with microscopic polyangiitis ILD (MPA-ILD) and 17 with ANCA-positive IP without systemic vasculitis (ANCA-IP). PPF occurred in 15/38 (39.5%) overall, and 27% of patients with MPA-ILD and 53% with ANCA-IP. Patient characteristics did not differ between PPF and non-PPF, however, the survival was significantly worse in patients with PPF than those with non-PPF. On multivariate regression analysis, higher age, higher serum SP-D level, and lower baseline %FVC were associated with PPF. In ANCA-ILD, 39.5% of patients demonstrated PPF, which is associated with increased mortality. Predictors of PPF were older age, higher SP-D, and lower baseline %FVC.


Data collection
Demographic, clinical, radiological, and PFT data including % predicted forced vital capacity (%FVC), and lung diffusing capacity for carbon monoxide (DL CO ) % predicted were retrieved from medical records.Data also included age, sex, obesity, disease duration, and smoking history (pack-years).Past and current treatment history were also documented.

Chest imaging
For all patients, HRCT images were acquired at diagnosis, regardless of respiratory symptoms.HRCT was performed pretreatment with follow up at 6 and/or 12 months thereafter.
All HRCT images were reviewed by 4 pulmonologists (S.H., K.S., A.S., and Y.U.) and 1 specialist pulmonary radiologist (A.K.) who were blinded to patient clinical and laboratory findings.Judgements were compared relative to clinical features and laboratory findings for each patient.All scans were acquired during inspiration with the patient supine; slice thickness was ≤ 1.0 mm at 10-20-mm intervals from the apices to the lung bases.IP was classified as usual interstitial pneumonia (UIP), probable UIP, indeterminate for UIP, and alternative diagnosis according to the 2022 updated ATS/ERS/JRS/ALAT PPF guidelines 18 .
We evaluated serial changes in HRCT findings before and at 6, 12, and 24 months.Using the PPF guidelines, disease progression was designated as the following findings on HRCT within the past year: (1) Increased extent or severity of traction bronchiectasis and bronchiectasis; (2) New ground-glass opacity with traction bronchiectasis: (3) New fine reticulation; (4) Increased extent or increased coarseness of reticular abnormality; (5) New or increased honeycombing; and (6) Increased lobar volume loss.

Pulmonary function tests
Lung volumes, forced expiratory volume in 1 s, and DL CO were measured by using the Chestac-8800 automated pulmonary testing system (Chest Co. Ltd., Tokyo, Japan) according to standard methods.Values were expressed as percentages of the predicted value.Data at baseline, 6 months, 1 year, 2 years, and last follow up were recorded.
Serial PFT trends at 1 year, expressed as percentages of baseline values, were evaluated for FVC.
In a patient with ILD of known or unknown etiology other than IPF and overt radiological evidence of pulmonary fibrosis, PPF is defined as at least two of the following three criteria occurring within the past 1 year with no alternative explanation: (1) Worsening respiratory symptoms; (2) Physiological evidence of disease progression defined as: (i) absolute decline in FVC ≥ 5% predicted within 1 year of follow-up or (ii) absolute decline in DL CO (corrected for Hb) ≥ 10% predicted within 1 year of follow-up; and (3) Radiological evidence of disease progression defined as: (a) increased extent or severity of traction bronchiectasis and bronchiolectasis; (b) new ground-glass opacity with traction bronchiectasis; (c) new fine reticulation; (d) increased extent or increased coarseness of reticular abnormality; (e) new or increased honeycombing; and (f) increased lobar volume loss 18 .
We compared PPF and non-PPF in 38 patients who underwent PFTs and were followed at least 1 year with comparison for patient background and prognosis/outcome for both groups.Additionally, multivariate analysis was performed to examine factors associated with PPF.
Finally, we performed a survival analysis to compare idiopathic interstitial pneumonias (IIPs) and ANCA-ILD, with similar HRCT patterns.Using data from the IIPs cohort at our institution, 63 patients with IPF and 79 with idiopathic nonspecific interstitial pneumonia (iNSIP) were identified and compared to those with ANCA-ILDs (ANCA-UIP, ANCA-NSIP).We also evaluated the incidence of PPF and compared this among these groups.

Statistical analysis
Continuous variables are expressed as median (range), unless otherwise stated; comparison was made by using the Mann-Whitney U test.Categorical variables were compared by using the χ2 test.A P value of < 0.05 was considered to indicate statistical significance.All statistical analyses were performed by using SPSS version 11.0 (SPSS Inc., Chicago, IL).

Ethical approval
This retrospective study was conducted in accordance with the amended Declaration of Helsinki and was approved by the Ethics Committee of Toho University Omori Medical Center in April 2022 (project approval number M20221).Considering the retrospective design, the requirement for informed consent was waived by Ethics Committee of Toho University Omori Medical Center because of anonymized patient data.The study protocol was in accordance with relevant guidelines.

Patient characteristics
In total, 56 patients with ANCA-ILD were identified and included.Of these, 38 patients (19 men and 19 women; mean age 72 years) underwent routine PFTs at intervals of over 1 year.Their characteristics are shown in Table 1.This patient cohort comprised 21 with MPA-ILD and 17 with interstitial pneumonia with ANCA positivity alone with no systemic vasculitis (ANCA-IP).The patients were then divided into MPA-ILD and ANCA-IP groups.Age, sex, and smoking history did not differ between MPA-ILD and ANCA-IP.Evaluation of both UIP and probable UIP patterns as UIP patterns showed a significantly higher proportion of UIP patterns for ANCA-IP than for MPA-ILD (65% vs 29%, P = 0.03).All MPA-ILD patients (n = 21) were treated with high-dose (> 0.6 mg/kg) glucocorticoids (GC).Of these, 16 received immunosuppressant therapy; only 1 received antifibrotic agents.In contrast, in the ANCA-IP group (n = 17), 10 patients were treated with low dose (≤ 0.5 mg/kg) GC, 2 were treated with antifibrotic agents, while 6 patients did not receive any treatment.

Laboratory investigations and pulmonary function test (PFT) results
Laboratory results for MPA-ILD and ANCA-IP are also shown in Table 1.Serum MPO-ANCA levels were significantly higher in the MPA-ILD compared to ANCA-IP group (375.3 vs. 71.7 U/mL, P = 0.006).Serum CRP level tended to be higher in MPA-ILD than ANCA-IP, but was not significant (5.1 vs. 1.8 mg/dL, P = 0.352).Levels of serum ILD biomarkers such as KL-6 and SP-D, did not differ between the two groups (948.5 vs. 992.4U/mL, P = 0.399; 190.0 vs. 172.7 ng/mL, P = 1.000).For PFTs, mean baseline FVC and %FVC did not differ between the two groups (2370 vs. 2550 mL P = 0.179, 88.6 vs. 92.5%,P = 0.383).Mean baseline %DL CO did not differ between the two groups (66.7 vs. 69.8%,P = 0.709).All MPA-ILD patients received high-dose GC and/ or immunosuppressants like cyclophosphamide.Also, various treatments were administered in the ANCA-IP group depending on the HRCT pattern and/or pathologic pattern including low-dose GC, immunosuppressants, and/or antifibrotic agents.Use of high-dose GC and immunosuppressant therapy was significantly higher in the MPA-ILD vs ANCA-IP group (100 vs 0%, P < 0.001; 76.2 vs. 5.9%, P = 0.002).Use of antifibrotic agents did not differ between both groups (4.8 vs. 11.8%,P = 0.577).The proportion of patients who received no treatment was significantly lower in MPA-ILD than in ANCA-IP (0 vs. 35.3%,P = 0.003).

Comparison of clinical features between PPF and non-PPF
Table 2 shows patient characteristics, laboratory findings, and PFT results for patients with PPF and non-PPF disease.Age, sex, and smoking history did not differ between the groups.Serum MPO-ANCA and CRP levels did not differ between the groups (55.2 vs. 327.7 U/mL, P = 0.26; 2.3 mg/dL vs. 4.2 mg/dL P = 0.36).Serum KL-6 and SP-D levels tended to be higher in PPF than non-PPF; the difference was not significant (1158 vs. 843 U/ mL, P = 0.425; 253.1 vs. 120.2ng/mL, P = 0.068).Regarding PFTs, FVC and %FVC tended to be lower in the PPF patients than non-PPF but the differences were not significant (2370 mL vs. 2550 mL, P = 0.145; 88.6% vs. 92.5%,P = 0.121).Also, %DL CO did not differ between the two groups (66.7% vs. 69.8%,P = 0.845).Together, no significant baseline clinical differences were demonstrated when comparing PPF and non-PPF patients.

Discussion
To our knowledge, a few previous studies have investigated serial changes in clinical symptomatology, PFT results, and HRCT findings in patients with MPA-ILD or ANCA-IP 20,21 .Our study found a relatively high incidence where approximately 40% of patients with ANCA-ILD demonstrated PPF.Among the ANCA-ILD cohort, PPF was more frequent in ANCA-IP than in MPA-ILD.www.nature.com/scientificreports/Cases with FVC decline over time and worse prognosis in IPF have been reported 1 .Around one-third of fILDs have a clinical disease behavior similar to IPF, demonstrating a PPF in those patients with IIPs, CTD-ILD, HP, and conditions other than IPF, and these patients have a worse prognosis.The concept of PF-ILD was proposed in the INBUILD trial 17 against this background and was defined as PPF in the updated 2022 IPF guideline 18 .The clinical entity of PPF describes a high-risk population in patients with fILDs and manifests as deterioration in pulmonary symptoms, HRCT findings, and pulmonary function.
We identified that older age, higher serum SP-D level, and lower %FVC at baseline were clinical predictors of PPF in ANCA-ILD.It has been shown that older age 22 , higher serum SP-D level 23 , and lower %FVC 24 at baseline were associated with worse outcome in patients with IPF.In another report, fibrosis HRCT pattern at baseline, diabetes mellitus, and steroid-use posed a higher risk of development of PF-ILD in patients with CTD-ILD 25 .www.nature.com/scientificreports/Furthermore, our study demonstrated an association of PPF with increased mortality in patients with ANCA-ILD.Our results are consistent with the previous INBUILD trial which identified patients with PF-ILD as a high-risk group for worse outcome 17 .In the INBUILD trial, UIP-like pattern on HRCT among patients with non-IPF fibrotic ILD was associated with greater FVC decline 17 .In this study, however, the proportion of UIP patterns on HRCT was significantly higher for ANCA-IP than for MPA-ILD; UIP pattern was not associated with PPF.These findings might be due to insufficient sample size, and we need to accumulate more cases to confirm these findings.Takakuwa et al. investigated survival and prognostic factors in patients with ANCA-ILD and found survival was significantly lower in patients with UIP pattern than those with the NSIP pattern 20 .According to Libra et al., evaluating the role of p-ANCA in predicting clinical evolution and prognosis in a cohort of IPF patients revealed similar lung function decline in IPF patients with and without p-ANCA during follow-up.However, IPF p-ANCA + showed better survival 21 .Similarly, our study showed that the survival was better in ANCA-UIP than IPF.These results indicate potential behavioral similarity of ANCA-UIP and UIP associated with collagen vascular disease.
Although, the efficacy of nintedanib in PF-ILD for patients with non-IPF disease was reported in the INBUILD trial, there is no similar evidence for ANCA-ILD.To date, the role of anti-inflammatory or anti-fibrotic agents in ANCA-ILD has not been evaluated.In our study, all MPA-ILD patients received high-dose corticosteroids and immunosuppressants such as cyclophosphamide.However, 27% of MPA-ILD patients developed PPF despite immunosuppressant therapy.Therefore, the efficacy of antifibrotic agents in patients with PPF despite use of anti-inflammatory agents in MPA-ILD needs to be examined.Nevertheless, the ANCA-IP cohort received various treatments determined by the specific HRCT and/or pathologic pattern.These medications included corticosteroids, immunosuppressants, and/or antifibrotic agents in spite of which 53% of ANCA-IP developed PPF.
To date, there is no established treatment protocol for patients with ANCA-IP.Several studies have demonstrated the role of immunosuppressants in reducing the risk of developing MPA in MPO-ANCA-positive patients initially diagnosed with IPF 11,12,25 .Specifically, patients who do not receive immunosuppressants might be at a relatively high risk of developing MPA.According to Hozumi et al., the absence of immunosuppressant or anti-fibrotic treatment for ANCA-IP was correlated with MPA 26 .
Hosoda et al. 27 reported significantly increased attenuation areas around honeycombing on chest HRCT in patients with UIP and MPO-ANCA positivity but without MPA (ANCA/UIP).Pathologically, ANCA/UIP is characterized by more prominent inflammatory cell infiltration, lymphoid follicles with germinal centers, and cellular bronchiolitis.Takemura et al. reported that in ANCA-IP with UIP HRCT pattern, patients showed pathological findings of cellular inflammation in interstitial tissue 28 , destructive bronchiolitis, and cysts and that these findings were observed radiologically as honeycombing with increased attenuation around honeycombing.These pathological findings suggest that anti-inflammatory treatment may be effective even in the UIP pattern and should be verified in the future.A prospective study is requisite to validate the clinical effectiveness of immunosuppressant or anti-fibrotic treatment for patients with ANCA-IP and PPF.
Novel developments in endotyping regarding biological understanding of disease and the emerging field of precision medicine have revealed the inadequacy of "a one-size-fits-all approach" in the comprehensive management of chronic lung diseases.Although several reports identified candidate diagnostic or prognostic biomarkers for IPF, impeding factors have hindered the translation of these results into clinical practice 29 .Our findings approximate a step for robust precision medicine approaches in pulmonary fibrosis, potentially highlighting the importance of ANCA endotyping towards an innovative approach in precision medicine.The utility of endotyping with ANCA as a prognostic or theragnostic biomarker in pulmonary fibrosis requires further investigation.
This study has limitations that warrant consideration.First, this was a single-center retrospective study with a small sample size.A large-scale multicenter study is needed to confirm these findings.Second, the study included various HRCT and/or pathologic patterns, which may inadvertently affect the treatment effect and prognosis.Finally, patients with ANCA-IP received various treatments based on HRCT and/or pathologic pattern such as corticosteroids, immunosuppressants, and/or antifibrotic agents.Some patients did not receive any treatment.This variation in treatment regimens may affect the results.

Conclusion
Among all patients with ANCA-ILD, 39.5% showed features consistent with PPF.PPF is associated with increased mortality in patients with ANCA-ILD.Older age, high SP-D level, and lower %FVC at baseline were associated with PPF in patients with ANCA-ILD.These findings highlight the need for follow-up assessment using PFTs and HRCT in these patients.Studies with a larger patient population and longer follow-up duration are needed to clarify the course of ILD in patients with ANCA-ILD.

Figure 4 .
Figure 4. Kaplan-Meier plots for survival analysis from time of diagnosis among PPF and non-PPF groups.

Figure 5 .
Figure 5. Kaplan-Meier plots for survival analysis from time of diagnosis divided into MPA-ILD and ANCA-IP among PPF and non-PPF groups.

Figure 6 .
Figure 6.Kaplan-Meier plots for survival analysis from time of diagnosis in idiopathic interstitial pneumonias and ANCA-ILD with similar HRCT pattern.(A) ANCA-UIP vs IPF.(B) ANCA-NSIP vs idiopathic NSIP.