Fibrotic interstitial lung diseases (ILD) encompass a wide spectrum of diseases, associated with respiratory symptoms, exercise limitation, impaired quality of life, and the potential progression to respiratory failure and death. Among them, idiopathic pulmonary fibrosis (IPF) is the most frequent and prototypical disease, associated with a risk of disease progression and a dismal prognosis [1]. Progressive pulmonary fibrosis (PPF) refers to ILDs other than IPF, harbouring a disease behaviour comparable to IPF, with progression of fibrosis despite appropriate management [2–4]. With some variation depending on the underlying ILD diagnosis, one third of patients with non-IPF ILDs develop PPF [5,6], a disease course which is associated with a poor outcome similar to that of IPF [4].

The antifibrotic drugs pirfenidone and nintedanib are approved worldwide for a treatment for IPF, and are generally used as monotherapy. Both drugs approximately halve the rate of annual decline in forced vital capacity (FVC) [7,8], and could reduce mortality and exacerbation rate [9]. Similarly, nintedanib has demonstrated efficacy and is approved in patients with PPF [1,10]. There is weaker evidence that pirfenidone may also slow disease progression in patients with PPF [10–12], and pirfenidone is not approved nor recommended in this indication [1]. In most cases however, antifibrotic monotherapy is not sufficient to halt disease progression in both IPF or PPF, highlighting the critical need for more effective drugs and therapeutic strategies.

Pirfenidone and nintedanib are synthetic molecules administered orally. There is evidence indicating that each drug has distinct mechanisms of action within the fibrotic cascade [13–15], providing a theoretical rationale for combining them to further reduce lung function decline [16,17]. Pharmacokinetic studies did not show relevant pirfenidone–nintedanib interactions [18,19]. However, their side effect profiles partially overlap, notably with digestive adverse events and asthenia. Even when used as monotherapy, their tolerability is often a limiting factor to long-term continuation, leading in real-world studies to dose reduction or even discontinuation [20–22], and potentially hampering the long-term benefits of treatment.

Experience of the use of these drugs in combination is limited to 3–6 months clinical trials, and to a few short-term observational studies [18,23–25]. Data on long-term safety and tolerability of combination therapy are limited. In this retrospective, multicentre, observational study, we aimed at assessing the tolerability and clinical outcomes of the combination of pirfenidone and nintedanib in patients with IPF or PPF treated in real-world setting, with a special focus on the subgroup of patients in whom the combination therapy was given as a bridge to lung transplantation.

We reported the outcomes of 38 patients treated with a combination of pirfenidone and nintedanib for IPF or PPF. Upon inclusion, patients exhibited a severe functional impairment (50.0% had a FVC<60%, and mean DLCO was 32% of predicted). Approximately half of the patients received first-line nintedanib, while the other half received pirfenidone. The median duration of initial monotherapy was 25.7 months [15.2; 33.6]. At the initiation of combination therapy, the add-on antifibrotic treatment was administered at full recommended dosages of nintedanib and pirfenidone, in 82.4% and 73.7% of cases respectively.

The safety profile of the combination therapy was consistent with that of the individual antifibrotic drugs. ADRs were observed in 84.2% of patients, and were severe in 28.9%. The most frequent ADRs were weight loss (loss of ≥5% of initial body weight in 52.6% of patients), diarrhoea (36.8%), abdominal pain and anorexia (28.9% each).

For monotherapy, data issued from clinical trial [7,8,26] and real-world studies [27–29] indicate that the incidence of adverse events for nintedanib and pirfenidone were, respectively 4–32% and 5–30% for weight loss, 24–63% and 3–25% for diarrhoea, and 8–29% and 12–17% for anorexia.

As compared to monotherapy, only limited data are available regarding combination therapy of pirfenidone and nintedanib in patients with IPF/PPF. Three clinical trials investigated the use of a combination of nintedanib and pirfenidone in patients with IPF [18,23,25], including one by Vancheri et al. in which patients treated with nintedanib were randomized to receive add-on pirfenidone for 12 weeks versus nintedanib alone [23]. In this study, nintedanib with add-on pirfenidone was not associated with an increased incidence of ADRs or serious ADRs (monotherapy vs combination: 89% vs. 88%, and 3.8% vs. 9.8%, respectively). However, similar to our study, ADRs were the primary reason for discontinuation of combination therapy. These trials did not report any worsening in quality of life among patients undergoing combination therapy [23,25].

Overall, previous trials and real-life studies on combination therapy reported an incidence of ADRs ranging from 71% to 89% (any ADR), with serious ADRs occurring in 4% to 7% of cases. The most common adverse events included diarrhoea (33–49%), anorexia (11–39%), consistent with our findings [18,23–25,30–32].

In our cohort, two patients (5.3%) experienced grade II+ liver enzyme elevation, in line with previous findings [31]. In both cases, the liver function normalized after the interruption of both antifibrotic agents, and both antifibrotics could be reintroduced sequentially, without recurrence of hepatic adverse event.

The incidence of weight loss in our cohort was higher than previously reported for combination therapy (52.6% vs. reported rates of 7–17%) [18,23–25,30–32], and was severe in 23.7%. This could be explained not only by the additive effect from the two antifibrotic combined, but also by the advanced stage of the disease in our patients [33,34]. Tolerability data may also differ between real-life studies and randomized controlled trials due to differences in the rate and nature of comorbidities [23,25], as well as in patient follow-up and management of adverse events. There is also some heterogeneity in the data reported and a lack of precision on the definition of weight loss across studies [31].

The safety profile of combination therapy in our cohort was otherwise consistent with findings from other studies, and in line with that observed with monotherapy. This could be attributed to a selection bias of patients, as patients with poor tolerance to nintedanib or pirfenidone monotherapy were likely not considered for combination therapy. Furthermore, nintedanib and pirfenidone seem to involve different metabolization pathways [35,36]. Data do not suggest significant drug-drug interactions between both antifibrotic treatments [19,23], aside from a potential yet controversial reduction in nintedanib bioavailability when co-administered with pirfenidone [18].

Overall continuation of combination therapy was relatively good, with patients remaining on treatment for a median duration of 12.8 months. Combination therapy was terminated in 26.3% of patients (excluding death or lung transplantation), dose was reduced in 28.9%, and dose was continued unchanged in 44.7% of patients. In the literature, treatment discontinuation rates of monotherapy are reported at 10–21%, dose modification rate of 15–26% and continuation without modification rate of 45–58% [7,8,27–29]. For combination therapy [23–25], previous studies reported rates of 15–49%, 27–43% and 30%, respectively.

Similar to the findings reported by Hisata et al. [24], we observed that the majority of treatment discontinuations for ADRs occurred during the first 12 months. Hisata suggested that physicians were inclined to stop either treatment at the first sign of toxicity, as questions remained about the efficacy and safety of combination therapy. As the first ADR generally occurs shortly after initiation of either nintedanib or pirfenidone [23,37], an early discontinuation of combination therapy would be expected.

Concerning the efficacy of combined treatment, we observed an attenuation in the decline in FVC and FEV1, consistent with previous observations [23–25]. As our study is observational, these results must be taken with caution. The PROGRESSION trial (NCT03939520), an ongoing phase IV trial, is currently evaluating the efficacy and tolerability of a combined therapy of nintedanib and pirfenidone, as compared to unchanged monotherapy, or a switch from one monotherapy to the other, in patients with IPF experiencing disease progression despite first-line antifibrotic therapy.

Antifibrotic drugs reduce the risk of exacerbations of IPF/PPF [9]. Here, we reported the number of exacerbations following discontinuation of combination therapy, however the methodology prevented to derive conclusions on changes in exacerbation rates pre- and post-combination therapy.

In the subgroup of patients listed for lung transplantation and receiving combination therapy, over half (11/21) underwent transplantation during follow-up. Five patients out of 21 (23.8%) discontinued combination therapy among the entire subgroup. We would have hypothesized a better persistence on treatment in the subgroup of patients listed for transplantation due to a possible higher motivation to maintain eligibility, but the rate of treatment cessation in this group was comparable to that of the entire cohort. Secondly, out of 5 patients who discontinued treatment, 4 experienced a >5% reduction in body weight from baseline. Combination therapy may have been discontinued to mitigate further loss of weight as patients prepared for lung transplantation. Interestingly, several studies reported that the continuation of antifibrotics until lung transplantation was safe [38,39].

The limitations of our study include its retrospective design, the limited statistical power due to the small sample size, the absence of comparative arm with monotherapy antifibrotic treatment, and the lack of detailed information about patients’ comorbidities or comedications, which might have influenced clinical outcomes and adverse events. However, the pre-and-post design of the exploratory efficacy analysis allows for a certain degree of comparison against monotherapy regimen. Moreover, since only patients who survived and were not transplanted contributed post-treatment spirometry, and many were already in advanced stages of disease, the apparent slowing of FVC (mL) decline may partly reflect survivor bias and a natural ‘floor effect’, rather than a treatment-specific benefit. As the study population was relatively young and included many patients awaiting lung transplantation, the findings may have limited generalisability to the broader IPF/PPF population.

Despite these limitations, our study has several strengths, including its multicentre, international design. There are limited data on antifibrotic combination therapy in PPF, and this study adds to the emerging evidence in this area.

Although no new safety signal arose, combination therapy is challenging in real-world setting due to poor tolerability especially weight loss. It can nevertheless be a viable treatment option in some patients, particularly as a bridge to lung transplantation. Further studies are needed to confirm the efficacy of this combined therapeutic strategy.

Combining pirfenidone and nintedanib is challenging in real-world setting due to poor tolerability especially weight loss. It can nevertheless be a viable treatment option in some patients, particularly as a bridge to lung transplantation.

Corentin Meersseman: Original Draft Preparation, Formal Analysis, Investigation, Resources, Data Curation, Review & Editing, Approval.

Elisa Martínez Besteiro: Data acquisition, Investigation, Review & Editing, Approval.

Nicolas Romain-Scelle: Data acquisition, Formal Analysis, Investigation, Review & Editing, Approval.

Bruno Crestani: Data acquisition, Investigation, Review & Editing, Approval.

Sylvain Marchand-Adam: Data acquisition, Investigation, Review & Editing, Approval.

Hilario Nunes: Data acquisition, Investigation, Review & Editing, Approval.

Lidwine Wémeau-Stervinou: Data acquisition, Investigation, Review & Editing, Approval.

Raphael Borie: Data acquisition, Investigation, Review & Editing, Approval.

Rémi Diesler: Data acquisition, Investigation, Review & Editing, Approval.

Claudia Valenzuela: Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Resources, Data Curation, Review & Editing, Visualization, Supervision.

Vincent Cottin: Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Resources, Data Curation, Writing, Review & Editing, Visualization, Supervision.

The data supporting the reported results can be obtained from the first author upon reasonable request.

Artificial intelligence was not used for this manuscript.

This work was not funded.

All authors have provided their potential conflicts of interest following the ICMJE guidelines in a file attached to this submission.

Data are available for sharing upon reasonable request.