Interstitial pneumonia with autoimmune features show better survival and less exacerbations compared to idiopathic pulmonary fibrosis

ILD patients who were consecutively enrolled at the time of diagnosis between 2008 and 2015 were evaluated. Clinical data was collected prospectively in this ILD cohort. Our institution has a longitudinal ILD cohort and the clinical and laboratory data of patients in this cohort have been collected prospectively using an ILD-pre-specified protocol. Databases were collected on a regular basis and in real-time at the time of work up. For the present study, data has been retrospectively reviewed, while radiologic imaging and pathologic findings were re-reviewed by radiologists and pathologists.

According to the European Respiratory Society/American Thoracic Society research statement on IPAF, 54 patients were categorized into the IPAF group [1]. All CTD-ILD patients had been referred to rheumatologists before diagnosis. Clinical characteristics and overall survival of the IPAF patients were compared to those of CTD-ILD patients and IPF patients. For comparison of clinical characteristics, the IPF group was further categorized into seronegative and seropositive IPF subgroups.

The diagnosis of IPAF was done following the European Respiratory Society/American Thoracic Society research statement [1]. The patients’ clinical data, laboratory, radiologic and histopathologic findings were evaluated to check whether they met the criteria of clinical, serologic and morphologic domains. After exclusion of other possible causes, such as malignancy or heart failure, initial HRCT images were checked for features of multi-compartment involvement such as unexplained pericardial effusion, and pleural effusion and thickening. Unexplained intrinsic airways disease was defined as never smokers with forced expiratory volume in one second/forced vital capacity (FEV1/FVC) ratio of < 70%. We defined it as such, as similar findings in ever-smokers would mean a possibility of concurrent chronic obstructive pulmonary disease [11]. Pulmonary vasculopathy was determined by the presence of pulmonary hypertension (PH) defined as a mean pulmonary artery pressure ≥ 25 mmHg via right heart catheterization (RHC) [1]. RHC is the gold standard diagnostic method for the diagnosis of PH, but we did not perform RHC routinely as this is an invasive procedure [12]. In the present study, findings coherent to high right ventricular load detected on trans-thoracic Doppler echocardiography [13] or FVC/diffusing capacity for carbon monoxide (DLco) > 1.6 and DLco < 60% were used as criteria for the definition of pulmonary hypertension [1, 3, 14]. In order to exclude the possibility of an overlap with CTD-ILD, all our potential IPAF patients were seen by rheumatologists.

AE of non-IPF ILD groups and IPF group were defined using the revised definition of IPF proposed in 2016 [15]. There is, however, no existing official definition of AE-ILD in non-IPF ILD [16]. Time duration between enrolment to the cohort and first ILD exacerbation was estimated. An exacerbation of ILD was defined as a patient being admitted to the hospital due to acute aggravation of ILD occuring less than 1 month before admission [15].

Patients with IPF who presented with positive features of the serologic domain, but without any features of the clinical and morphologic domains of IPAF were classified as the seropositive IPF. Patients with IPF who did not have positive features of the serologic domain of IPAF were classified as seronegative IPF.

This retrospective study was approved by the Institutional Review Board of Bucheon St Mary’s Hospital. The need for informed consent was waived because the study was a retrospective review.

Among 305 patients evaluated, 54 (17.7%) patients met the classification of IPAF. Table 1 shows detailed features within each domain of IPAF patients. A total of 17 (31.5%) patients showed features coherent with the criteria of the clinical domain. Arthralgia was the most common symptom (76.5%), followed by Raynaud’s phenomenon and unexplained digital oedema (both 16.7%). Among 54 IPAF patients, 49 (90.7%) patients met the criteria of the serologic domain. ANA abnormality was the most common finding (63.3%), followed by high rheumatoid factor levels and positive anti-cyclic citrullinated peptide (anti-CCP) (14 (28.6%) and 7 (14.3%) patients, respectively). A total of 44 (81.5%) patients met the morphologic domain criteria. Among the 39 patients with features of the morphologic domain, 34 (87.2%) patients showed NSIP patterns on HRCT findings, which was the most common ILD pattern in the IPAF group. Three (7.7%) patients presented with organizing pneumonia. Among 34 patients who underwent diagnostic biopsy, 12 patients showed histopathologic patterns consistent with features of the morphologic domain. In 5 patients, interstitial lymphoid aggregates with germinal centres were the most common finding (41.7%) among patients with histopathologic patterns related to IPAF, followed by pathologic findings consistent with organising pneumonia (4 patients, 33.3%). Findings defined as multi-compartment involvement were seen in 11 (20.4%) patients; unexplained pleural effusion or thickening in 5 (45.5%) patients, unexplained pericardial effusion or thickening in 3 (27.3%) patients, unexplained intrinsic airways diseases in 2 (18.2%) patients and unexplained pulmonary vasculopathy in 1 (9.1%) patient.

Clinical characteristics of the IPAF patients were compared to 76 CTD-ILD, 145 seronegative IPF and 30 seropositive IPF (Table 2) patients. The CTD-ILD group was comprised of 46 patients with rheumatoid arthritis (RA), 18 patients with systemic sclerosis, 6 patients with Sjogren’s syndrome, 5 patients with inflammatory muscle disease and 1 patient with systemic lupus erythematous (SLE).

The proportion of male (35.2%) patients was significantly lower in the IPAF group than in both the seronegative and seropositive IPF groups (p < 0.001, 71.0 and 73.3%, respectively), but similar to that of the CTD-ILD (31.2%) group. Mean age and the proportion of ever smokers was significantly lower in both the IPAF and CTD-ILD groups than in the seronegative and seropositive IPF (p < 0.001 and p < 0.001, respectively) groups.

There was a significant difference between the groups for ILD patterns observed on the initial HRCT (p < 0.001). In the IPF group, 100% of the patients presented with UIP patterns on HRCT and no radiological NSIP pattern was noted. Usual interstitial pneumonia (UIP) pattern on the initial HRCT was seen in 25.9% of the IPAF group and 46.1% of the CTD-ILD group. The NSIP pattern was the most frequent ILD pattern observed in the IPAF group (63%). While only 22.4% of the CTD-ILD group showed the NSIP pattern, which was the second most frequent pattern after UIP. Emphysematous change was seen in only 9.3% of the IPAF group, which is lower than in the CTD-ILD group, and the seropositive and seronegative IPF groups (p = 0.006).

During the observation period, AE-ILD was seen in 25.9% of the IPAF group, 32.9% of the CTD-ILD and 35.4% of the IPF group (p < 0.001). The proportions of patients who experienced exacerbations were significantly different in 1, 3 and 5 years after enrolment in the cohort (p = 0.022, p = 0.026 and p = 0.007, respectively). When time to first exacerbation was compared between the groups, the IPAF group showed no significant difference compared to the IPF group, while the CTD-ILD group showed a significantly longer duration compared to the IPF group (p = 0.02). Mean value of time to first exacerbation was 29.5 ± 27.5 months in the IPAF group, 32.6 ± 29.7 months in the CTD-ILD group and 17.3 ± 21.4 months in the IPF group (Table 3).

OS was compared between the different combinations of ILD types. Mean survival time was 73.3 months in the IPAF group, 104.0 months in the CTD-ILD, 50.7 months in the seronegative IPF group, and 56.5 months in the seropositive IPF group. Difference on OS among the four groups was statistically significant (p < 0.001) (Fig. 1). When the seronegative and seropositive IPF groups were combined into a single group, a significant difference in OS of the three groups was seen as well (p < 0.001) (Table 3) (Fig. 2), with a mean survival time of 52.0 months in the IPF group. When the IPAF patients with UIP pattern (n = 15) were compared to the IPF group, OS was slightly better in the IPAF than in the IPF group (64.6 months vs 52.0 months, respectively), with no significant difference (p = 0.08) (Fig. 3).

For analysis of association with mortality, the following variables - age, gender, smoking history, autoimmune antibody positivity, first line treatment, emphysema from HRCT, FVC, DLCO, whether patients experienced exacerbation and ILD type, were entered in the univariate analysis. Age, gender, smoking history, autoimmune antibody positivity, FVC, DLCO, ILD exacerbation and ILD type were significant factors from the univariate analysis, and were entered in the multivariate analysis. Age, FVC, ILD exacerbation and ILD type were significant risk factors for mortality in the multivariate analysis (p = 0.034, HR 1.022, 95% CI: 1.002–1.044; p < 0.001, HR 0.970, 95% CI: 0.955–0.984; p = 0.001, HR 2.074, 95% CI: 1.366–3.148; p = 0.047, HR 0.436, 95% CI: 0.192–0.984 (IPAF vs IPF), respectively). From the multivariate analysis, the IPAF type was associated with significantly better survival when compared to the IPF (Table 4).