2485: Prevalence of progressive pulmonary fibrosis in systemic sclerosis-associated interstitial lung disease

Background/Purpose: To estimate the prevalence of the progressive pulmonary fibrosis (PPF) phenotype among patients with systemic sclerosis-associated interstitial lung disease (SSc-ILD) and identify risk factors that may predict its development.

Methods: A multicentric ambispective study was conducted in four tertiary Spanish hospitals. The diagnosis of PPF was based on the 2022 ATS/ERS/JRS/ALAT criteria for progressive pulmonary fibrosis, requiring at least two of the following three criteria: (1) increased fibrosis on HRCT; (2) worsening respiratory symptoms; and (3) an absolute decline in predicted FVC% by ≥5% or in DLCO% by ≥10% within one year of follow-up

Results: The series included 168 patients with SSc-ILD confirmed by thoracic HRCT. All patients fulfilled the 2013 ACR/EULAR classification criteria for SSc. Based on radiological findings, 70.8% (119) had an NSIP pattern, 11.3% (19) a UIP pattern, 3.6% (6) an OP pattern, 5.4% (9) mixed patterns, and 8.9% (15) an indeterminate pattern.

Of the 168 patients, 37 (22.0%) exhibited radiological worsening, 27 (16.1%) experienced clinical worsening, and 37 (22.0%) showed functional decline. Of these, 31 (18.5%) met the criteria for PPF.

In the comparative analysis, PPF patients were more often men (39% vs. 11.4%; p=0.001) and had a higher prevalence of oesophageal involvement (89.3% vs. 63.6%; p=0.010). Pulmonary arterial hypertension was also significantly more frequent (32.1% vs. 12.5%; p=0.023).

The median time from ILD diagnosis to the evaluation visit was similar between groups: 81.5±84 months (IQR 47–180) in patients with PPF and 71±73 months (IQR 25–131) in those without PPF (p=0.126). PPF patients had lower baseline %pFVC (69.1±18 vs. 88.5±24; p< 0.001) and %pDLCO (51.8±14 vs. 62.9±17; p=0.004), and a higher proportion of %pFVC < 70 (57.1% vs. 19.3%; p=0.001) and %pDLCO < 60 (64.3% vs. 19.3%; p=0.001). Their 6MWT distance was also shorter (356±97 m vs. 409±102 m; p=0.028). As expected, lung transplantation was exclusive to the PPF group (42.8%; p< 0.0001), and mortality was significantly higher (14.3% vs. 2.3%; p=0.014), despite the greater use of biologic agents as rescue therapy (37.3% vs. 0.7%; p=0.001).

A multivariable logistic regression analysis with Ridge penalisation was performed to assess predictors of PPF. The model included six variables: age at diagnosis, oesophageal involvement, PAH, symptomatic ILD, %pFVC < 70, and %pDLCO < 60 (see Table 2). Younger age at diagnosis (coefficient = –1.72), oesophageal involvement (+0.17), PAH (+0.10), and %pFVC < 70 (+0.17) were positively associated with PPF development, whereas symptomatic ILD (+0.08) and %pDLCO < 60 (–0.17) showed weaker associations (See Table 1).

Conclusion: In our series, the prevalence of PPF in SSc-ILD reaches 18.5%. Factors such as %pFVC < 70 at ILD diagnosis, esophageal involvement, and the development of PAH appear to increase the risk of PPF.