Survival and lung function decline in patients with definite, probable and possible idiopathic pulmonary fibrosis treated with pirfenidone.

BACKGROUND: There is no clear evidence whether pirfenidone has a benefit in patients with probable or possible UIP, i.e. when idiopathic pulmonary fibrosis (IPF) is diagnosed with a lower degree of diagnostic certainty. We report on outcomes of treatment with pirfenidone in IPF patients diagnosed with various degrees of certainty. METHODS AND
FINDINGS: We followed patients in the multi-national European MultiPartner IPF Registry (EMPIRE) first seen between 2015 and 2018. Patients were assessed with HRCT, histopathology and received a multi-disciplinary team (MDT) IPF diagnosis. Endpoints of interest were overall survival (OS), progression-free survival (PFS) and lung function decline.
RESULTS: A total of 1626 patients were analysed, treated with either pirfenidone (N = 808) or receiving no antifibrotic treatment (N = 818). When patients treated with pirfenidone were compared to patients not receiving antifibrotic treatment, OS (one-, two- and three-year probability of survival 0.871 vs 0.798; 0.728 vs 0.632; 0.579 vs 0.556, P = 0.002), and PFS (one-, two- and three-year probability of survival 0.597 vs 0.536; 0.309 vs 0.281; 0.158 vs 0.148, P = 0.043) was higher, and FVC decline smaller (-0.073 l/yr vs -0.169 l/yr, P = 0.017). The benefit of pirfenidone on OS and PFS was also seen in patients with probable or possible IPF.
CONCLUSIONS: This EMPIRE analysis confirms the favourable outcomes observed for pirfenidone treatment in patients with definitive IPF and indicates benefits also for patients with probable or possible IPF.

Introduction

Idiopathic pulmonary fibrosis (IPF) is characterized by progressive fibro-proliferative scarring of the lung parenchyma, originating from alveolar lesions. IPF is most frequently seen in predisposed middle-aged and older individuals, ultimately leading to respiratory failure and death.

Despite current treatment options with antifibrotics, the prognosis of IPF remains unfavourable. Two drugs, namely pirfenidone and nintedanib, have become available in the last decade, and have demonstrated to slow progression of IPF. However, antifibrotic treatment cannot reverse fibrotic lung remodelling.

Randomised clinical trials (RCT), followed by open extensions of these trials and post-hoc assessments, have documented treatment efficacy for these two drugs [1-8]. Patients participating in these RCTs were rather homogeneous with regard to age, comorbidity pattern and high-resolution computed tomography (HRCT) findings. As a consequence, real-world IPF patients very often do not match inclusion criteria of these RCTs [9]. One group of particular concern is IPF patients who do not present with definite usual interstitial pneumonia (UIP) on HRCT, but demonstrate the radiological and/or histopathological pattern of probable/ possible UIP.

The RCTs, which have led to the registration of pirfenidone, included either 1) patients with a definite UIP pattern on HRCT or 2) patients with only a probable/possible UIP on HRCT, but later confirmation of UIP by lung biopsy [10]. Despite the many studies, which have investigated pirfenidone, there is still lack of evidence as to whether pirfenidone can influence outcomes in patients with a probable/possible UIP pattern on HRCT, and no subsequent histopathological confirmation of UIP. In many elderly patients presenting with probable or possible UIP, the risk for diagnostic lung biopsy is often considered too high, thereby preventing possibly beneficial antifibrotic IPF treatment.

Real-world registries can be a valuable source of information to demonstrate the effectiveness of pirfenidone for these particular subgroups of patients [11, 12]. Furthermore, length of follow-up of RCTs is limited. When IPF registries run over several years, they can, in addition to RCTs, provide survival data for IPF subgroups for longer time periods.

Publications based on IPF real-world registries have recently described the epidemiology, the clinical management, and the treatment outcomes associated with IPF in various countries [13-18]. Although IPF registries usually follow a greater number of participants than RCTs, they can still be limited, because of small numbers of patients falling into specific sub-categories such as probable or possible UIP.

At the time of reporting, more than 3,500 IPF patients from 50 hospitals in eleven European and Middle Eastern countries have been enrolled into the EMPIRE registry [14, 19–21]. Based on this large and multi-national IPF population, important new evidence for small subgroups of IPF patients can be generated.

In the present analysis, we report overall survival, progression-free survival and lung function decline for patients treated with pirfenidone according to subgroups of different diagnostic certainty as defined by HRCT pattern and histopathology.

Methods

Study population

This analysis includes IPF patients with their first visit to the EMPIRE registry taking place between 1 January 2015 and 31 December 2018 and followed through 29 October 2019.

IPF was diagnosed according to the 2011 ATS/ERS/JRS/ALAT criteria [22]. That is, all patients included in the EMPIRE registry were considered to have had HRCT, lung biopsy for histopathological assessment if indicated, and a MDT discussion for final IPF diagnosis. Final IPF diagnosis was always determined by a local multidisciplinary team (MDT); the EMPIRE data therefore also include patients with HRCT or histopathological findings not demonstrating UIP pattern.

The present data analysis included 1) patients treated with pirfenidone and 2) patients not treated with antifibrotics (herein referred as the no antifibrotic treatment group) and was done according to subgroups of diagnostic certainty (see below). Patients followed in EMPIRE and receiving nintedanib at any time were excluded from analysis (see S1 Fig).

For the purpose of longitudinal analysis, a participant’s baseline visit (start of follow-up) was defined as follows: 1) patients treated with pirfenidone were analysed from the time onwards when treatment was started; 2) patients on no antifibrotic treatment were analysed from the time when the first visit to EMPIRE was recorded. End of follow-up was defined when death, lung transplantation or progression of disease (for PFS analysis) was observed, and was censored either at the date of the last visit to the registry, or the date of the last documented pirfenidone treatment (for the pirfenidone group) or the date when a patient was lost to follow-up. If a patient terminated the pirfenidone therapy, the subsequent treatment period (no antifibrotics) was not included in the analysis.

Study measurements (forced vital capacity–FVC, diffusing capacity for carbon monoxide–DLCO, and six-minute walk test– 6MWT) were collected at baseline and approximately every 6 months thereafter. However, the frequency of visits might fluctuate in accordance with local clinical practice or regulatory requirements for the monitoring of pirfenidone therapy.

All measurements were entered in a standardised way into a web-based data management tool with secure electronic data transfer. All data was cross-checked for plausibility (using in-built algorithms during data entry) and outlying observations. The study was approved by the Ethics Committee of Thomayer University Hospital and Institute for Clinical and Experimental Medicine, Prague, Czech Republic. The EMPIRE registry was approved by local ethics committees in individual countries and sites involved in the registry. All patients signed an informed consent prior to enrolment into the registry.

EMPIRE was set-up to reflect the real-world management of patients with IPF. IPF patients were enrolled by hospitals in European and Middle Eastern countries with a long-standing expertise in the management of interstitial lung disease (ILDs). Given the expertise of hospitals participating in EMPIRE, external central radiological or histopathological review were not considered necessary. IPF patients were always enrolled into EMPIRE at the discretion of the treating physician, but only when the diagnostic work-up including HRCT, lung biopsy if indicated, and MDT diagnosis was completed.

At baseline, participating IPF patients were assessed with HRCT to define UIP, possible UIP, and inconsistent with UIP; and histopathology, if indicated, to define UIP, probable UIP, possible UIP, and absence of (= no) UIP. Based on the combination of HRCT and lung biopsy findings, the final IPF diagnosis (definitive IPF, probable/ possible IPF, not IPF) was determined. The combination of HRCT and lung biopsy findings defined the subgroups for analysis (see Table 1).

Table 1

Baseline characteristics of patients with pirfenidone and patients on no antifibrotic treatment.

Data are given as mean (±SD) or N (%).

		Total N = 1626	Pirfenidone N = 808	No antifibrotic treatment N = 818	P
Demographics	Men	1 153 (70.9%)	609 (75.4%)	544 (66.5%)	< 0.001
	Age (years)	68.6 (±9.6)	67.7 (±8.8)	69.5 (±10.2)	< 0.001
	BMI	28.1 (±4.4)	28.4 (±4.4)	27.7 (±4.3)	0.001
Smoking	Never-smokers	598 (36.9%)	289 (35.8%)	309 (38.0%)	0.066
	Ex-smokers	227 (14.0%)	101 (12.5%)	126 (15.5%)
	Current smokers	796 (49.1%)	418 (51.7%)	378 (46.5%)
HRCT pattern	Definite UIP	1 096 (67.4%)	582 (72.0%)	514 (62.8%)	< 0.001
	Possible UIP	453 (27.9%)	191 (23.6%)	262 (32.0%)
	Inconsistent with UIP	70 (4.3%)	35 (4.3%)	35 (4.3%)
	Not performed	7 (0.4%)	0 (0.0%)	7 (0.9%)
Histopathology	UIP	161 (9.9%)	111 (13.7%)	50 (6.1%)	< 0.001
	Probable UIP	64 (3.9%)	41 (5.1%)	23 (2.8%)
	Possible UIP	58 (3.6%)	32 (4.0%)	26 (3.2%)
	Not UIP	38 (2.3%)	14 (1.7%)	24 (2.9%)
	Not performed	1 305 (80.3%)	610 (75.5%)	695 (85.0%)
IPF diagnosis	IPF	1 162 (71.5%)	627 (77.6%)	535 (65.4%)	< 0.001
	Probable + possible IPF	368 (22.6%)	140 (17.3%)	228 (27.9%)
	Not IPF	89 (5.5%)	41 (5.1%)	48 (5.9%)
	Not performed	7 (0.4%)	0 (0.0%)	7 (0.9%)
Comorbidities	Number of comorbidities	3.41 (1.99)	3.58 (2.01)	3.24 (1.95)	< 0.001
	Heart and vascular	1 150 (70.7%)	596 (73.8%)	554 (67.7%)	0.007
	Pulmonary	483 (29.7%)	251 (31.1%)	232 (28.4%)	0.233
	Gastrointestinal	863 (53.1%)	483 (59.8%)	380 (46.5%)	< 0.001
	Urogenital	259 (15.9%)	133 (16.5%)	126 (15.4%)	0.560
	Cancer	94 (5.8%)	45 (5.6%)	49 (6.0%)	0.716
IPF treatment	Pharmacological	1 018 (64.1%)	808 (100.0%)	210 (26.9%)	< 0.001
	Rehabilitation	256 (16.2%)	191 (23.7%)	65 (8.4%)	< 0.001
	LTOT	299 (18.9%)	170 (21.1%)	129 (16.6%)	0.022
	Lung transplantation	146 (9.2%)	103 (12.8%)	43 (5.5%)	< 0.001
Lung functions at baseline ± 3 months	FVC predicted (%)	79.4 (±21.5) / 1,2741	73.9 (±16.1) / 5531	83.6 (±24.0) / 7211	< 0.001
	DLCO predicted (%)	49.2 (±19.1) / 1,1881	46.7 (±14.5) / 5181	51.2 (±21.8) / 6701	0.008
GAP index	I	581 (45.9%)	258 (41.6%)	323 (50.1%)	0.008
	II	553 (43.7%)	289 (46.6%)	264 (40.9%)
	III	131 (10.4%)	73 (11.8%)	58 (9.0%)
Dyspnoea	NYHA I	100 (8.7%)	29 (4.7%)	71 (13.2%)	< 0.001
	NYHA II	597 (51.8%)	335 (54.6%)	262 (48.6%)
	NYHA III	432 (37.5%)	236 (38.4%)	196 (36.4%)
	NYHA IV	24 (2.1%)	14 (2.3%)	10 (1.9%)

1 Number of patients for whom the baseline value of FVC predicted or DLCO predicted was available

Outcomes

We report outcomes (OS, PFS and FVC decline) for the pirfenidone group and the no antifibrotic treatment group, and compare patients according to subgroups of diagnostic certainty. Short-term (lung function decline) and long-term (OS, PFS) outcomes of treatment were evaluated. Short-term outcomes included the change of FVC and DLCO for the first 12 months of follow-up. Long-term outcomes included the progression of disease or death from any cause. Progression of disease was defined present, when either a decline of FVC > 10%, or a decline of DLCO > 15%, or a decline of 6MWD > 50 m was observed, whichever came first, compared to baseline values. Progression free survival (PFS) and overall survival (OS) was calculated accordingly. Other parameters analysed at baseline only were GAP index [23] and dyspnoea rated according to the NYHA criteria [24].

Statistical analysis

For continuous variables we report mean and standard deviation and for categorical variables absolute and relative frequency.

Kaplan-Meier methodology was used for the analysis of OS and PFS. Patients with no observed event during follow-up (death, lung transplantation or progression of disease, see definition above) were censored at the date of the last visit to the registry, the date of last documented pirfenidone treatment or the date of loss to follow-up, whichever came last.

To adjust for potential confounding, we used multivariate Cox proportional hazard models including covariates likely associated with outcomes OS and PFS. These covariates (age, sex, height, FVC at baseline and dyspnoea) were included into all Cox proportional hazard models. Using these Cox models, we then tested for interaction between pirfenidone (treatment vs no treatment) and the categories of diagnostic certainty (definitive, probable, possible).

Lung function in the first 12 months decline was analysed using a linear, mixed effects model. In this analysis we included only patients with a minimum of six months of follow up. The model on lung function decline was adjusted for age, height, sex, FVC at baseline and dyspnoea. The annual lung function decline was described by estimates of the time-dependent slope (including 95% CI). A random intercept and random time slope were included in the model specification. The model also tested for the significance of interaction between pirfenidone treatment and categories of diagnostic certainty. All statistical analysis was carried out using SPSS 25.0.0.0 and STATA 14.2. The level of significance α was set at 0.05.

Results

Patient characteristics at baseline

At baseline 1,626 IPF patients were included; 808 patients were treated with pirfenidone and 818 patients received no antifibrotic treatment. There was a higher frequency of visits in the patients treated with pirfenidone than in the patients without antifibrotic treatment (mean 3.3 vs 1.9 visits for the first 12 months of follow-up, and 5.3 vs 2.7 visits for the first 36 months of follow-up, respectively). The pirfenidone group included more men (75.4% vs 66.5%), was slightly younger (mean age 67.7 vs 69.5 years), included a smaller percentage of non-smokers (35.8% vs 38.0%), had higher body mass index (28.4 vs 27.7), and a less favourable GAP index (GAP 1: 41.6% vs. 50.1%) and dyspnoea distribution (NYHA I: 29 (4.7%) vs. 71 (13.2%); NYHA I + II: 59.3% vs 61.8%) at baseline. Further, patients in the pirfenidone group had lower FVC (absolute FVC mean 2.47 l vs 2.71 l; predicted FVC mean 73.9% vs 83.6%) and DLCO (predicted mean 46.7 vs 51.2%) at baseline.

The most frequently reported comorbidities were cardiovascular (73.8% vs 67.7% of patients), gastrointestinal (59.8% vs 46.5%) and pulmonary (31.1% vs 28.4%) for the comparison between the pirfenidone and the no antifibrotic treatment group.

In patients treated with pirfenidone and stratified according to the diagnostic certainty of IPF, the proportion of men was highest in the group presenting with an UIP pattern on HRCT. Subgroups of different diagnostic certainty did not substantially differ with regard to FVC, DLCO, GAP index and dyspnoea.

Baseline characteristics of patients are presented in Table 1 (all patients) and in S1 Table (patients treated with pirfenidone stratified according to the certainty of IPF diagnosis). Due to the small number of patients with possible IPF according to the 2011 diagnostic criteria (7 in the pirfenidone group and 3 in the group with no antifibrotic treatment), the patients with possible IPF were combined with patients with probable IPF in data analysis.

Overall survival and progression-free survival

In patients with IPF and regardless of the degree of diagnostic certainty, OS was greater in patients treated with pirfenidone compared to those not receiving antifibrotic treatment. For the two groups (pirfenidone vs no antifibrotic treatment) survival was 87.1% vs 79.8% after one year, 72.8% vs 63.2% after two years, and 57.9% vs 55.6% after three years of follow-up respectively (Fig 1A).

Fig 1

Overall survival (a) and progression-free survival (b) in the whole IPF cohort.

A similar pattern was observed for PFS. PFS for the two groups (pirfenidone vs no antifibrotic treatment) was 59.7% vs 53.6% after one year, 30.9% vs 28.1% after two years, and 15.8% vs 14.5% after three years of follow-up, respectively (Fig 1B). Median time to progression of IPF was 15.3 months for the pirfenidone and 13.9 months for the no antifibrotic treatment group.

In multivariate Cox proportional hazard models, co-variates sex, FVC and dyspnoea at baseline were found significantly associated with both OS and PFS (S2 and S3 Tables). All covariates were kept in following models to adjust for potential confounding. For adjusted analysis, treatment with pirfenidone was significantly associated with better OS (HR 0.749, 95% CI 0.575 to 0.976; P = 0.032), indicating a 25.1% reduction in mortality. However, no association was observed when PFS was analysed following adjustment (1.037, 95% CI 0.880 to 1.222; P = 0.666), see Table 2.

Table 2

Hazard ratios associated with mortality and progression of IPF comparing pirfenidone with no antifibrotic treatment according to diagnostic subgroups (adjusted for age, sex, height, NYHA and FVC at baseline).

		No. of patients	Mortality HR (95% CI)			Progression HR (95% CI)
				P1	P for interaction2		P1	P for interaction2
Total		1,128	0.749 (0.575; 0.976)	0.032		1.037 (0.880; 1.222)	0.666
HRCT	UIP	735	0.668 (0.492; 0.906)	0.010	0.579	0.845 (0.696; 1.027)	0.090	0.109
	Possible UIP	342	0.765 (0.465; 1.259)	0.292		1.198 (0.892; 1.610)	0.229
	Inconsistent with UIP	51	0.418 (0.150; 1.161)	0.094		0.749 (0.386; 1.452)	0.392
Histopathology	UIP	119	0.495 (0.218; 1.122)	0.092	0.578	0.647 (0.369; 1.134)	0.129	0.497
	Probable UIP	52	0.449 (0.140; 1.436)	0.177		0.624 (0.306; 1.272)	0.194
	Possible UIP	40	0.345 (0.088; 1.347)	0.126		0.618 (0.275; 1.387)	0.243
	Not UIP	28	-	-		1.560 (0.530; 4.589)	0.419
IPF diagnosis	IPF	787	0.653 (0.486; 0.876)	0.005	0.401	0.850 (0.703; 1.028)	0.094	0.194
	Probable + possible IPF	267	0.798 (0.462; 1.378)	0.418		1.189 (0.859; 1.646)	0.297
	No	74	0.335 (0.106; 1.053)	0.061		0.985 (0.531; 1.829)	0.963

1 difference pirfenidone vs no antifibrotic treatment

2 differential effect of pirfenidone between diagnostic subgroups; higher P value (above 0.05) indicates that the possible effect of pirfenidone therapy (a difference between pirfenidone and no antifibrotic treatment) is similar across all diagnostic subgroups

Overall survival and progression-free survival according to diagnostic certainty of IPF

Overall survival was higher in patients treated with pirfenidone in all subgroups according to the diagnostic certainty (statistically significant results were observed namely for HRCT–UIP and final IPF diagnosis). Our analysis with multivariate model did not demonstrate a different association between pirfenidone treatment and OS for the different diagnostic IPF subgroups, i.e. definitive, and probable/possible IPF (Table 2, P value for interaction > 0.40 for each of the three classifications); although there seems to be a less pronounced difference in crude survival curves for some subgroups (S2 Fig). OS in patients treated with pirfenidone was similar across all diagnostic subgroups (S3 Fig).

Although the unadjusted comparison suggested a possible effect of pirfenidone on PFS (S4 Fig), also observed among subgroups (S5 Fig), this was not confirmed by adjusted multivariate analysis (Table 2).

Lung function decline in first 12 months

The FVC decline was -0.073 l/year (95% CI, -0.124 to -0.023 l/year) in the pirfenidone group and -0.169 l/year (95% CI, -0.230 to -0.109 l/year) in the group not receiving antifibrotic treatment (P = 0.017) (Table 3). On the other hand, no significant difference was observed for DLCO decline: -0.187 mmol/kPa/min (95% CI, -0.396 to -0.023 mmol/kPa/min) in the pirfenidone group, -0.116 mmol/kPa/min (95% CI, -0.355 to -0.122 mmol/kPa/min) in the no antifibrotic treatment group (P = 0.665) (S4 Table).

Table 3

Difference of annual FVC decline according to diagnostic subgroups (adjusted for age, sex, height, NYHA and absolute FVC at baseline).

		No. of patients (pirfenidone)	FVC decline annual rate–pirfenidone (95% CI) (l/yr)	No. of patients (no antifibrotics)	FVC decline annual rate–no antifibrotic treatment (95% CI) (l/yr)	P1	P value for interaction2
Total		526	-0.073 (-0.124; -0.023)	454	-0.169 (-0.230; -0.109)	0.017
HRCT	UIP	361	-0.078 (-0.140; -0.016)	275	-0.209 (-0.289; -0.130)	0.011	0.047
	Possible UIP	142	-0.040 (-0.137; 0.057)	158	-0.138 (-0.236; -0.041)	0.160
	Inconsistent with UIP	23	-0.181 (-0.410; 0.048)	21	0.061 (-0.212; 0.334)	0.183
Histopathology	UIP	71	-0.095 (-0.210; 0.021)	28	-0.163 (-0.374; 0.047)	0.575	0.178
	Probable UIP	28	-0.110 (-0.293; 0.073)	15	0.138 (-0.111; 0.388)	0.115
	Possible UIP	23	0.017 (-0.178; 0.212)	14	-0.144 (-0.369; 0.081)	0.288
	Not UIP	9	0.086 (-0.238; 0.411)	17	-0.116 (-0.336; 0.104)	0.312
IPF diagnosis	IPF	393	-0.073 (-0.132; -0.015)	285	-0.193 (-0.272; -0.114)	0.017	0.513
	Probable + possible IPF	101	-0.078 (-0.191; 0.034)	144	-0.167 (-0.272; -0.062)	0.261
	Not IPF	32	-0.072 (-0.300; 0.155)	25	-0.003 (-0.213; 0.208)	0.660

1 difference pirfenidone vs no antifibrotic treatment

2 differential effect of pirfenidone between diagnostic subgroups; higher P value (above 0.05) indicates that the possible effect of pirfenidone therapy (a difference between pirfenidone and no antifibrotic treatment) is similar across all diagnostic subgroups

Lung function decline in first 12 months according to diagnostic certainty of IPF

The effect of pirfenidone to slow FVC decline was most significant in patients with a UIP pattern on HRCT (-0.078 l/yr for the pirfenidone group, 95% CI, -0.140 to -0.016 l/yr; -0.209 l/yr for the no antifibrotic treatment group, 95% CI, -0.289 to -0.130 l/yr; P = 0.011) and in patients with confirmed IPF (-0.073 l/yr for the pirfenidone group, 95% CI -0.132 to -0.015 l/yr; -0.193 l/yr for the no antifibrotic treatment group; 95% CI -0.272 to -0.114; P = 0.017) (Table 3). We found a borderline statistically significant difference in FVC decline according to HRCT subgroups and in relation to pirfenidone treatment (P = 0.047), suggesting a potentially stronger effect of pirfenidone in patients with definite UIP/IPF.

No significant difference was observed for DLCO decline within diagnostic subgroups, irrespective of treatment with pirfenidone (S4 Table).

Discussion

Using follow-up data from the multi-national EMPIRE registry, we compared diagnostic subgroups of IPF patients as defined by the 2011 ATS/ERS/JRS/ALAT criteria. The clinical characteristics and course of IPF for patients receiving pirfenidone were compared to those of patients on no antifibrotic treatment. Our results are in line with randomized clinical trials and other real-world data, strongly indicating that patients with IPF profit from pirfenidone, experiencing longer OS and smaller FVC decline [1, 2, 6, 7].

Our results further suggest that the effect of pirfenidone on these outcomes was not significantly different between subgroups of IPF as defined by diagnostic certainty. Patients treated with pirfenidone in EMPIRE were found similar to other real-life cohorts as reported in literature (in terms of age and sex) However, FVC % predicted at baseline was rather lower compared to other real-life studies [13, 15, 17, 25–27].

In the literature, OS for patients treated with pirfenidone varies with 1-year OS between 85% and 99% [15, 16, 25], 2-year OS between 76.9% and 83% [16, 17], and 3-year OS between 73% and 74% [15, 26]. The OS in EMPIRE were similar or slightly lower (87.1%, 72.8% and 57.9% for one-, two, and three-year follow-up). Median time to death was therefore not reached during 36-month follow-up.

Differences in progression-free survival for the pirfenidone group and the no antifibrotic treatment group were not as apparent as for the overall survival. The Kaplan-Meier analysis indicates an effect lasting approximately 12 months. A more pronounced impact of antifibrotic treatment on OS rather than on lung function decline was recently reported from the German IPF registry study [18]. A higher frequency of visits in the pirfenidone group may also have increased the probability of detecting progression of IPF earlier.

Nevertheless, some IPF patients were lost to follow-up, the proportion being markedly higher for the group receiving no antifibrotic treatment, likely resulting into more favourable effect estimates for the group with no antifibrotic treatment, than would have been observed provided all the patients stayed in the study [12]. This differential loss to follow-up could have caused underestimation of the difference for PFS and lung function later on, when mainly patients with better health status at baseline remained alive. This was also illustrated in the comparison of baseline characteristics of patients completing 24 months of follow-up [28]. The median time to progression of patients treated with pirfenidone in our study is comparable with data published in literature [16, 27].

The effect of pirfenidone on OS was most clearly seen in the group of patients with the highest diagnostic certainty, i.e. those with a radiological and/or histopathological pattern of definite UIP and/or a final IPF diagnosis. This finding can be attributed to the much higher number of patients in this groups compared to IPF patients with a lower diagnostic certainty. Nevertheless, our analysis indicates that the effect of pirfenidone is not significantly different between the diagnostic IPF subgroups, suggesting that patients with probable and possible UIP/IPF might also profit from the pirfenidone treatment.

To our knowledge, this is the first registry-based study including real-world patients specifically focusing on differences in treatment outcomes related to the degree of diagnostic certainty of IPF. Long-term follow-up of the large real-life EMPIRE cohort allowed us to illustrate these relationships; however, quality of data needs to be taken into account when interpreting the data. The EMPIRE data have been collected in nearly 50 centres from 11 countries; this sampling could have caused increased between-site variability, and therefore lead to underestimation of study results. Furthermore, given that this is a patient registry and not a randomised clinical trial, a number of sources of bias and uncertainties should be considered (healthy participation bias, different baseline patient characteristics, rules for drug administration, different time order of diagnosis, admission visit and treatment initiation, etc.).

Some limitations of this analysis based on data from the EMPIRE registry might arise from its real-life nature and multicentre data collection. In the majority of cases, diagnosis of IPF is based on the assessment of HRCT images, which can be subjective. IPF diagnosis was determined in each participating centre without central reading or re-assessment. Although the patients with definite UIP/IPF prevail in the analysed cohort, there is also a sufficiently large cohort of patients with HRCT findings of lower certainty (probable and possible UIP) available for the comparison.

Management of IPF in clinical practice faces several limitations arising from disease characteristics and real-world settings: IPF is a rare disease and there is significant loss to follow-up of patients with less favourable prognosis and baseline characteristics, particularly when they do not receive antifibrotics. These patients are less prone to travel and usually receive palliative care at home provided by their general practitioners. On the other hand, patients with better baseline characteristics remain under follow-up for longer time periods, which may lead to the conclusion that health status has improved over time (e.g. in terms of greater FVC or less dyspnoea reported). This type of bias was illustrated in our previous study, in which baseline characteristics of patients staying at risk (under follow-up) for 6, 12, 18, and 24 months were compared–numbers of patients at risk decreased over time and the survivors were characterised by a higher FVC and DLCO, and less advanced dyspnoea at baseline, particularly in the group on no antifibrotic treatment [28]. The effect of drop-out results in generally low numbers of patients available for analysis not so much at diagnosis or treatment initiation, but mainly after a longer follow-up period (e.g. 3 years and more), when the remaining group of patients may have different characteristics than the original cohort due to substantial and selective drop out [12]. The real-world IPF studies published to date are usually based on dozens or lower hundreds of patients and follow-up is two or three years [13, 15–18, 25, 27], although some long-term data from large national or international IPF registries are available as well [29, 30].

It may be challenging to define a reasonable baseline visit in a retrospective registry-based study. Patients may be included in the registry at different times after diagnosis (or even before), and before or after treatment is started (both antifibrotic and non-antifibrotic). In this study, the baseline was set at pirfenidone therapy initiation for the pirfenidone group and admission visit (enrolment) for the no antifibrotic treatment group. The pirfenidone group and the no antifibrotic treatment group did not differ significantly in terms of the time pattern from diagnosis over admission visit to treatment initiation, although the analysed cohort included patients diagnosed before the defined period for enrolment (2015–2018). Mean and median time from diagnosis to admission visit was 9.81 and 0.41 months for the pirfenidone group and 13.57 and 0.82 months for the no antifibrotic treatment group, respectively. Mean and median time from the admission visit to the pirfenidone therapy initiation was 2.52 and 1.16 months, respectively.

In conclusion, this analysis of real-life data from the international EMPIRE registry confirmed favourable clinical outcomes associated with pirfenidone treatment. Treatment with pirfenidone was associated increased OS and a slower deterioration of lung function. This benefit of pirfenidone is likely present in IPF patients irrespective of their degree of diagnostic certainty.

Flow diagram of patients participating in the study and their availability for performed analyses.

(PDF)

Click here for additional data file.

Overall survival in diagnostic subgroups.

(PDF)

Click here for additional data file.

Overall survival of patients treated with pirfenidone in diagnostic subgroups.

(PDF)

Click here for additional data file.

Progression-free survival in diagnostic subgroups.

(PDF)

Click here for additional data file.

Progression-free survival of patients treated with pirfenidone in diagnostic subgroups.

(PDF)

Click here for additional data file.

Dataset for Fig 1.

(XLSX)

Click here for additional data file.

Abstract from the ATS conference.

Pirfenidone Effectiveness In Idiopathic Pulmonary Fibrosis With Different Radiologic Patterns.

(PDF)

Click here for additional data file.

Baseline characteristics of patients treated with pirfenidone according to diagnostic subgroups.

(PDF)

Click here for additional data file.

Univariate hazard ratio for mortality and PFS.

(PDF)

Click here for additional data file.

Multivariate hazard ratio for mortality and PFS (adjusted on age, sex, height, NYHA and FVC at baseline).

(PDF)

Click here for additional data file.

Difference of annual rate of decline in DLCO according to diagnostic subgroups.

(PDF)

Click here for additional data file.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

16 Feb 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Minghua Wu, M.D., Ph.D.

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/fileid=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf".

2. Thank you for stating the following financial disclosure:

“This study was supported by F. Hofmann-La Roche. The EMPIRE registry is supported by Boehringer Ingelheim and F. Hofmann-La Roche.”

Please state what role the funders took in the study.  If the funders had no role, please state: ""The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.""

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

3. Thank you for stating the following in the Acknowledgments Section of your manuscript:

“This study was supported by F. Hofmann-La Roche. The EMPIRE registry is supported by Boehringer Ingelheim and F. Hofmann-La Roche.

The authors thank investigators in clinical centres involved in the EMPIRE project for their valuable collaboration and data acquisition.”

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“This study was supported by F. Hofmann-La Roche. The EMPIRE registry is supported by Boehringer Ingelheim and F. Hofmann-La Roche.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

4. Thank you for stating the following in the Competing Interests section:

“Dr. Kirchgässler is an employee of F. Hoffmann-La Roche Ltd.

Dr. Lewandowska reports grants, personal fees and non-financial support from F. Hofmann-La Roche, grants, personal fees and non-financial support from Boehringer Ingelheim.

Dr. Stoeva reports and received fees from Boehringer Ingelheim for participation in advisory boards and from Roche for presentations.

Dr. Studnicka reports grants and personal fees from Boehringer-Ingelheim, grants and personal fees from Roche, during the conduct of the study.

Dr. Tekavec-Trkanjec reports personal fees from Roche, personal fees and other from Boehringer Ingelheim.”

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: ""This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

5. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere. [Some of the results were presented in a form of electronic poster at the ATS Congress 2020.] Please clarify whether this [conference proceeding or publication] was peer-reviewed and formally published. If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

6. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study focuses on the outcomes of the pirfenidone-treated group and the untreated group registered in the EMPIRE Registry. The study also compared the definite IPF with probable/possible UIP, and examined the differences in prognosis and changes in lung function depending on the degree of certainty of the UIP. The pirfenidone-treated group had better OS and PFS than the untreated group, but the multivariate results were significantly better only for OS, which was consistent with the previous report.

The authors further showed that the effect of pirfenidone is the same as that of UIP even with probable UIP pattern.

These results are consistent with previous studies, and since probable UIP is considered to be UIP in the current guidelines, examining the effect of pirfenidone treatment depending on the degree of UIP is a less important issue.

However, as real-world data, it is commendable that the pirfenidone-treated group had a better prognosis despite significantly lower baseline lung function than the untreated group.

Since we do not have to pay attention to the degree of UIP pattern in the antifibrotic treatment, it is a proposal to reconsider the title of this paper.

1) Since probable/possible UIP are diagnosed as UIP in the latest guideline, the title of this manuscript is of little appeal to physicians managing IPF treatment.

2) In this study, how are patients who have been discontinued from pirfenidone treatment or who have been switched from nintedanib to pirfenidone?

3) mention more deeply the difference from nintedanib treatment and how the results of this study affect when you are confused with nintedanib. There is a need to

4) Since nintedanib is indicated for PF-ILD, the chances of introducing pirfenidone are decreasing currently. The significance of the results of this study on the treatment of pirfenidone in such situations should be discussed more.

5) Do “other treatments” in Tables and Figures refer to “no antifibrotic treatment”? Does it also include nintedanib treatment as other treatment? It's easy to get confused, so it's better to unify them so that they are easy to understand.

6) Please describe the difference in the onset of acute exacerbations between the pirfenidone group and the untreated group. It's very important to discuss why the survival curves are different between these groups in this study. It is plausible that the prolongation effect of OS cannot be explained only by suppressing the decline in FVC by pirfenidone treatment.

7) In the section of the results, “Overall survival and progression-free survival according to diagnostic certainly of IPF”, the authors should explain the all data of the Figures in detail.

You mentioned that there was no significant different effect of pirfenidone treatment on OS for the different diagnostic IPF subgroups in Table 2 and Sup. Fig. 2. In Table 2, it seems that there was a difference between diagnostic certainly of IPF distinct from Sup Fig. 2. It was hard to understand your results for general readers.

Further, you described the OS in patients treated with pirfenidone regarding Sup. Fig. 3. However, Sup. Fig. 3 demonstrates PFS, but not OS, in diagnostic subgroups.

Moreover, I cannot find Sup. Fig. 5 in the supplementary file.

Please specify the definition of p1 and p2 in the Tables. Your presentation throughout the results including table capture is unfriendly to the readers.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Yasuhiko Koga

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

24 Mar 2022

Dear Dr. Minghua Wu,

We would like to thank you and the reviewer for the comprehensive review of our article. We have made changes to reflect readability and clarity of the paper as suggested.

Editors' comments

We modified and transferred information about funding and acknowledgements as requested. We explained connection to the previous publication (ATS poster) in the cover letter.

Please note that a new Conflict of Interest statement of Veronika Müller has been added.

Reviewers' comments

1) Since probable/possible UIP are diagnosed as UIP in the latest guideline, the title of this manuscript is of little appeal to physicians managing IPF treatment.

We agree that the stratification according to the diagnostic certainty with regard to applied treatment has currently become of lower importance. However, our paper demonstrates that this practice is justifiable and that the therapy likely works in all groups. Such evidence has been missing for pirfenidone. We therefore prefer to maintain the current title.

2) In this study, how are patients who have been discontinued from pirfenidone treatment or who have been switched from nintedanib to pirfenidone?

Described in Methods: “Patients followed in EMPIRE and receiving nintedanib at any time were excluded from analysis (see S1 Fig 1).” The “control” group contains only patients with no history of antifibrotic treatment. Figure legends and tables were corrected from “other treatment” to “no antifibrotic treatment” to be consistent with the rest of the paper.

3) mention more deeply the difference from nintedanib treatment and how the results of this study affect when you are confused with nintedanib. There is a need to

If the reviewer refers to switch from/to nintedanib, we agree that the nintedanib phase should not be included. This is why we excluded patients with previous nintedanib therapy from the analysis (since the final therapy outcome might be affected).

4) Since nintedanib is indicated for PF-ILD, the chances of introducing pirfenidone are decreasing currently. The significance of the results of this study on the treatment of pirfenidone in such situations should be discussed more.

This is a correct point, but we feel that it falls out of scope of our paper a bit. Furthermore, prescription of both antifibrotics differs from country to country, even within the EMPIRE registry; we therefore assume that pirfenidone remains an important option, at least for patients who do not tolerate the nintedanib therapy well.

5) Do “other treatments” in Tables and Figures refer to “no antifibrotic treatment”? Does it also include nintedanib treatment as other treatment? It’s easy to get confused, so it’s better to unify them so that they are easy to understand.

Yes, thank you, legends have been corrected as explained above.

6) Please describe the difference in the onset of acute exacerbations between the pirfenidone group and the untreated group. It’s very important to discuss why the survival curves are different between these groups in this study. It is plausible that the prolongation effect of OS cannot be explained only by suppressing the decline in FVC by pirfenidone treatment.

This is a reasonable assumption, but we found no difference in the onset of acute exacerbations between the pirfenidone and no antifibrotics group when analysing this dataset (median time to the first AE 5.38 months for PIR vs 5.48 months for no AF, P = 0.509). However, overall frequency of exacerbations was relatively low (60 events in the entire cohort of 1,600 patients).

7) In the section of the results, “Overall survival and progression-free survival according to diagnostic certainly of IPF”, the authors should explain all data of the Figures in detail.

We have added a brief interpretation and summary under each Figure in the supplementary material.

You mentioned that there was no significant different effect of pirfenidone treatment on OS for the different diagnostic IPF subgroups in Table 2 and Sup. Fig. 2. In Table 2, it seems that there was a difference between diagnostic certainly of IPF distinct from Sup Fig. 2. It was hard to understand your results for general readers.

Hopefully the interpretation under the supplementary figures helps. We also modified the main text referring to these outputs – Table 2 is an adjusted analysis in which differences between pirfenidone and no antifibrotics are assessed, while Suppl. Fig. 3 (originally Suppl. Fig. 2 in the reviewed version of the paper, see below) is an unadjusted analysis comparing pirfenidone patients only, without any involvement of patients with no antifibrotics.

Further, you described the OS in patients treated with pirfenidone regarding Sup. Fig. 3. However, Sup. Fig. 3 demonstrates PFS, but not OS, in diagnostic subgroups.

Moreover, I cannot find Sup. Fig. 5 in the supplementary file.

Apology for this inconvenience – the author responsible for the preparation of figures and supplementary material forgot to include Figure 1 in the supporting info document (a flow diagram of patients included in the analysis). Numbering of all supplementary figures therefore moved by one. The revised version includes the correct supplementary Figure 1 and numbering has been corrected.

Please specify the definition of p1 and p2 in the Tables. Your presentation throughout the results including table capture is unfriendly to the readers.

Added in headings and footnotes for better clarity.

Submitted filename: answers-reviewers.docx

Click here for additional data file.

10 Jun 2022

Submitted filename: revisions PLOS.docx

Click here for additional data file.

24 Jun 2022

We would like to thank the reviewer for the comprehensive review of our article. We included the suggested corrections and changes, and we think that the overall readability and clarity of the paper has been significantly improved.

Please see our answers and explanations below.

Sincerely

Martina Koziar Vašáková

On behalf of all co-authors

Revisions

1. Methods: “Patients with no observed event during follow-up (death, lung transplantation or progression of disease, see definition above) were censored at the date of the last visit to the registry, the date of last documented pirfenidone treatment or the date of loss to follow-up, whichever came last.” - It would be of value to know how patients that were started on pirfenidone and had to discontinue the medication early due to side effects were handled in the analysis. Eg. if patient was on pirfenidone for 1 month and then had to discontinue, was this patient still included in the pirfenidone arm? If there was a large proportion of patients like this, it would affect the interpretation of the results. Recommended addressing if this was considered and how it was handled, consider including the median duration of therapy (treatment with pirfenidone).

Patients treated with pirfenidone were always censored at the date of the PIR therapy termination. Our statement suggested that they were included in K-M even after this event, which was confusing. Thank you for this point, the text has been corrected accordingly.

2. Table 1: report standard deviations with � 

Added

3. Table 1: BMI and % of non-smokers reported in table do not match with what is written in the body of text.

Corrected in text, thank you

4. Table 1: in the FVC and DLCO category, does the number in parenthesis represents standard deviation? If so, consider including the �  to make it clear. Currently, it is confusing because in the first column the % sign is in parenthesis (%).

Added

5. Table 1: under smoking, is the P value representing the comparison between non-smokers? If so, then it should be placed in line with this row and not spanning the 3 rows. Similarly for HRCT pattern, Histopathology, IPF diagnosis, GAP index and dyspnoea.

The P value refers to the overall proportion of all categories of the particular variable in a patient group. E. g. Never/ex/current smokers, or NYHA I, II, III, IV. The patient groups may therefore be similar e.g. in the proportion of NYHA III or IV (the difference would be statistically insignificant), but the stages I and II make the statistically significant difference between PIR and No AF.

6. Table 1: baseline FVC and DLCO do not match with the data on supplementary Table 4.

Since a standalone output was published at ERS 2021 focused on healthy survivor bias, we decided to add this reference (no. 28) and remove Supplementary Table 4. There is slightly different methodological approach there in comparison to this paper and it might introduce confusion as pointed out by the reviewer. This also answers to the item 11.

7. Supplementary fig 1 flow diagram has an error. The numbers under patients with pirfenidone therapy and IPF diagnosis do not add up to 808 and do not match with Table 1. Table 1 does contain appropriate numbers as they add up to n=808.

Corrected, thank you

8. Table 2 and Table 3: caption 2 is confusing and misleading. It reads: “higher P value (above 0.05) indicates that the difference between pirfenidone and no antifibrotic treatment is consistent across all diagnostic subgroups,” this implies that there is a difference between the subgroups when the P value is above 0.05). Consider: higher P value (above 0.05) indicates that the difference between pirfenidone and no antifibrotic treatment in the different subgroups is not statistically significant.

We actually do not refer to the difference between PIR and NoAF alone, but this P value indicates whether this difference is similar across the diagnostic groups. If this P value was lower than 0.05, it would mean that pirfenidone has stronger effect in some of the diagnostic groups than in another. This happened in FVC decline with a borderline statistical significance (Table 3 – HRCT). But we agree that this particular output is not easy to understand – we have tried to make the explanation under Tables 2 and 3 more clear.

9. Supp. Table 3. The total number of patients on pirfenidone and on no antifibrotic treatment are different compared to the ones on Table 3 (FVC decline), why is this? Did some patients not have DLCO recorded?

Yes, completeness of lung function data is different and one of the biggest issues that we have to handle when analyzing data from our registry

10. Discussion, lines 271-272: For clarification, in which study was median time to death not reached?

In none of them, the same as in our study (all over 50% after 3-y f-up). Since the sentence might be confusing, we have changed it to “Median time to death was therefore not reached during 36-month follow-up.” (in our study)

11. Discussion, lines 312-314: is there a reference for this statement?

Addressed in item 6

12. Figure 2. The legend has been corrected to describe the groups as “pirfenidone group” and the “no antifibrotic treatment group”; however, under each individual image within figure 2, legend still reads PRI and OT. Same applies for Figure 1a,1b as well as S1 Fig 4. If keeping the abbreviation OT, please specify that it represents the no antifibrotic group.

Corrected to “No AF”, thank you

Grammatical/syntax corrections

Thank you, we have made the indicated corrections.

1. Methods: We followed patients in the multi-national European MultiPartner IPF Registry (EMPIRE) first seen in between 2015 and 2018.” – grammatical error, just remove “in” so that it reads first seen between 2015 and 2018.

2. Introduction: “One group of particular concern is IPF patients, who do not present with definite usual interstitial pneumonia (UIP) on HRCT, but demonstrate the radiological and/or histopathological pattern of probable/ possible UIP” - syntax error, remove comma after IPF patients

3. Results – Overall survival and progression-free survival according to diagnostic certainty of IPF: “although there seems to be less pronounced difference in crude survival curves for some subgroups (S1 Fig 2).” – this statement is missing an “a”- “although there seems to be a less pronounced difference in crude survival curves for some subgroups in the unadjusted survival analysis (S1 Fig 2).”

4. Results – Overall survival and progression-free survival according to diagnostic certainty of IPF: “Although the unadjusted comparison suggested a possible effect of pirfenidone on PFS (S1 Fig 4), with PFS in patients treated with pirfenidone rather similar among subgroups (S1 Fig 5), this was not confirmed by adjusted multivariate analysis (Table 2)” – This phrase can be made friendlier to the reader, consider -Although the unadjusted comparison suggested an a possible effect of pirfenidone on PFS (S1 Fig 4), also observed among subgroups (S1 Fig 5), this was not confirmed by adjusted multivariate analysis

5. Supplementary fig 1. Patients on/not on pirfenidone (or nintedanib) or patients not on pirfenidone (instead of with and without)

6. Line 87: should read: population, importance new evidence

7. Line 94: should read: January 1st 2015 and December 31st 2018 and followed through October 29th 2019.

8. Line 96: should read: patient included in the EMPIRE registry

9. Line 98: should read: indicated, and a MDT discussion for final…

10. Line 108: remove word “onwards”

11. Line 163: significance was set “at” instead of set “on”

12. Line 196: authors use a hyphen (-) in no-antifibrotic treatment. No hyphen is used throughout the rest of the article. Consider removing it to stay consistent throughout the text.

13. Line 258-260: “Patient’s characteristics and the course of IPF were analyzed for those, who were treated with pirfenidone and compared to those receiving no antifibrotic treatment.” – this sentence does not read well. Consider: The clinical characteristics and course of IPF for patients receiving pirfenidone were compared to those of patients on no antifibrotic treatment.

14. Line 318: decrease in time �  decrease over time

15. Line 320: patient with no antifibrotic treatment �  patients on no antifibrotic treatment

16. Fig 1 demographics instead of demography

Submitted filename: response-to-reviewers-2.docx

Click here for additional data file.

17 Aug 2022

Survival and lung function decline in patients with definite, probable and possible idiopathic pulmonary fibrosis treated with pirfenidone

PONE-D-21-27861R2

Dear Dr. Vasakova,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

James West, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: To the Authors,

The authors revised all of the comments of their manuscript.

All comments have been addressed

It is suitable for the publication of the PlosOne.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: YASUHIKO KOGA

Reviewer #2: No

**********

22 Aug 2022

PONE-D-21-27861R2

Survival and lung function decline in patients with definite, probable and possible idiopathic pulmonary fibrosis treated with pirfenidone

Dear Dr. Vasakova:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. James West

Academic Editor

PLOS ONE