Abstract
Introduction Interstitial lung diseases (ILDs) can be caused by mutations in the SFTPA1 and SFTPA2 genes, which encode the surfactant protein (SP) complex SP-A. Only 11 SFTPA1 or SFTPA2 mutations have so far been reported worldwide, of which five have been functionally assessed. In the framework of ILD molecular diagnosis, we identified 14 independent patients with pathogenic SFTPA1 or SFTPA2 mutations. The present study aimed to functionally assess the 11 different mutations identified and to accurately describe the disease phenotype of the patients and their affected relatives.
Methods The consequences of the 11 SFTPA1 or SFTPA2 mutations were analysed both in vitro, by studying the production and secretion of the corresponding mutated proteins and ex vivo, by analysing SP-A expression in lung tissue samples. The associated disease phenotypes were documented.
Results For the 11 identified mutations, protein production was preserved but secretion was abolished. The expression pattern of lung SP-A available in six patients was altered and the family history reported ILD and/or lung adenocarcinoma in 13 out of 14 families (93%). Among the 28 SFTPA1 or SFTPA2 mutation carriers, the mean age at ILD onset was 45 years (range 0.6–65 years) and 48% underwent lung transplantation (mean age 51 years). Seven carriers were asymptomatic.
Discussion This study, which expands the molecular and clinical spectrum of SP-A disorders, shows that pathogenic SFTPA1 or SFTPA2 mutations share similar consequences for SP-A secretion in cell models and in lung tissue immunostaining, whereas they are associated with a highly variable phenotypic expression of disease, ranging from severe forms requiring lung transplantation to incomplete penetrance.
Abstract
SFTPA1 and SFTPA2 mutations lead to similar alterations in SP-A secretion and lung tissue expression. They are associated with a highly variable phenotypic expression ranging from incomplete penetrance to severe interstitial lung diseases and lung cancer. https://bit.ly/30SrEVb
Introduction
Chronic interstitial lung diseases (ILDs) include a heterogeneous group of diffuse parenchymal lung disorders that can affect patients of all ages [1–3]. In adult patients, the most severe form of ILD is idiopathic pulmonary fibrosis (IPF), which is also the most frequent. The clinical course of ILD patients is highly variable and unpredictable [4–6]. An unknown part of this heterogeneity implicates genetic factors [7–9]. Pathogenic mutations have been identified in approximately 30% of familial ILD, mainly in genes encoding the telomerase complex in adult patients and mainly in genes encoding proteins of surfactant metabolism (SFTPA1, SFTPA2, SFTPB, SFTPC, ABCA3 and NKX2-1) in paediatric cases [7, 8, 10–13].
Information on SFTPA1 and SFTPA2 defects in ILD remains limited. These genes are closely related paralogs containing six exons each. They encode the surfactant proteins (SPs) SP-A1 and SP-A2, two collectins that oligomerise in a “flower bouquet” octadecamer to generate the SP-A complex. SP-A plays a structural role in the formation of the surfactant tubular myelin and also an immuno-modulatory role through its carbohydrate recognition domain (CRD) that is encoded by the last exon (i.e. exon 6) of SFTPA1 or SFTPA2. The molecular epidemiology of SFTPA1 and SFTPA2 is still largely unknown. Indeed, over the last few years, only 11 SFTPA1 or SFTPA2 mutations have been reported in patients with familial forms of ILD and lung cancer. The pathogenicity of these mutations was attested by functional tests for only five of them [11, 14–20].
In the present study, performed in the framework of ILD molecular diagnosis, we identified 11 missense SFTPA1 (n=3) or SFTPA2 (n=8) class 4 (likely pathogenic) and class 5 (pathogenic) mutations (according to the American College of Medical Genetics (ACMG) classification of sequence variations) in 14 independent probands [21]. These findings prompted us to assess the functional consequences of these missense variations by means of in vitro studies and to accurately describe the associated disease phenotypes in the 14 involved families.
Patients and Methods
Patients
Patients with an ILD and with no telomerase complex mutations were referred to us in the frame of molecular diagnosis. ILD diagnosis was assessed by expert clinicians at one of the 23 adult centres of the French reference network for rare lung diseases (RespiFIL) that covers the whole of France. Phenotypic information was collected including age at ILD onset, associated extra-respiratory features, lung cancer association and family history. Available high-resolution computed tomography (HRCT) scans and histologic samples were centrally reviewed [3].
When a class 4 or class 5 mutation was identified, genetic counselling was performed and a familial genetic screening was proposed to the symptomatic individuals and the asymptomatic major relatives [21]. Each family was discussed within the French multidisciplinary team meeting for genetic ILD [22]. The study was approved by the relevant ethics committees (the Comité de Protection des Personnes) and written informed consent was obtained from all participants or their legal representatives. Clinical information was collected in a legally authorised database (CNIL No. 681248).
Molecular analyses
SFTPA1 and SFTPA2 molecular analyses were performed on DNA extracted from peripheral blood leukocytes. The NM_005411.5 and NM_001098668.2 isoforms were used as a reference for SFTPA1 and SFTPA2, respectively. Coding exons and flanking intronic junctions of SFTPA1 and SFTPA2 were sequenced with a custom-targeted capture panel (SeqCap EZ Choice, Roche Sequencing, Pleasanton, CA, USA). The library was prepared following manufacturer's instructions. Data were analysed through an in-house double pipeline based on the Bowtie2 [23] and BWA [24] tools. Reads were visualised with the IGV viewer (Broad Institute, http://software.broadinstitute.org/software/igv/). Copy number variation analysis was performed with a depth-ratio comparison between subjects sequenced in the same run. Class 4 (n=10) and class 5 (n=1) mutations were checked by Sanger sequencing using the previously described protocol [19].
Haplotype analyses
Haplotype analyses for families 5, 6 and 11–14 were performed by combining SFTPA2 and flanking SFTPA1 single nucleotide polymorphism (SNP) genotyping by targeted-capture in probands and Sanger sequencing in probands and relatives (SFTPA1 is the closest gene to SFTPA2, 50.5kb downstream on chromosome 10). Phasing was performed through allele segregation within each family and through proband next generation sequencing (NGS) data when contiguous variations were included in at least four common sequencing reads.
Structural and electrostatic changes induced by the SP-A1 and SP-A2 mutations on the carbohydrate recognition domain
Three dimensional (3D) structures of the wild-type (WT) and mutant CRDs of SP-A1 and SP-A2 were generated using the crystal structure (residues 201–248) of rat SP-A obtained from the Protein Data Bank (ID: 3PAR; www.rcsb.org/structure/3PAR) as a template. Briefly, homology modelling was performed using Modeller software version 9.10 (https://salilab.org/modeller). Accuracy of output structures was further assessed using Procheck version 3.5.4 [25, 26]. Conformational changes were evaluated by a root-mean-square deviation (RMSD) value comparison between the WT and mutated SP-A1/SP-A2 domains using SuperPose (Back Bone) version 1.0 (http://superpose.wishartlab.com/help.html) [27]. An RMSD value of less than two is in favour of a very similar protein structure. Electrostatic modelling of the protein domain surface was generated with the use of PyMOL vacuum electrostatics version 0.99 (https://pymol.org/2).
Plasmid constructs
The WT SFTPA1 and WT SFTPA2 complementary DNA obtained from human universal RNA were inserted into the pcDNA3.1_V5_His_TOPO expression vector (Invitrogen, Carlsbad, CA, USA). A nine amino acid haemagglutinin (HA) tag and an eight amino acid FLAG tag were subsequently added after the signal peptide to generate the pSFTPA1_WT and pSFTPA2_WT plasmids, respectively. Site-directed mutagenesis was performed to generate the mutant plasmids carrying the identified mutations using a Quick Change Site Mutation Mutagenesis kit (E0554S, New England Biolabs, Ipswich, MA, USA). The resulting plasmid constructs were checked by Sanger sequencing of the inserts and cloning site flanking junctions.
Cell culture and transfection
WT and mutant pSFTPA1 and pSFTPA2 expression plasmids were used to assess protein production (cell lysate) and protein secretion (medium) after transfection in HEK293T cells. As such, 600 000 HEK293T cells were first cultured at 37 °C with 5% carbon dioxide in 35 mm wells in complete medium (comprising DMEM (1X)+GlutaMAX-I (4.5 g·L−1 d-glucose) from Life Technologies, Paisley, UK (31966-021); 10% fetal bovine serum (FBS); 1% penicillin and streptomycin). On Day 2, at 80% confluence, the cells were transfected with 1 µg of either pSFTPA1_WT, pSFTPA2_WT, the corresponding mutated plasmid, or the empty vector using the FuGENE HD (E2312, Promega, Madison, WI, USA) method in a 4:1 FuGENE:DNA ratio. On Day 4, the cells were re-fed with complete medium without FBS. Protein production was studied on Day 5.
Western blot analyses
Aliquots of the cellular lysates and of the cell culture medium were analysed by Western blot. The membranes were incubated overnight at 4 °C, with either a monoclonal anti-HA peroxidase high-affinity antibody (12013819001, Roche, Mannheim, Germany; 1:500 ratio) or with a monoclonal anti-FLAG peroxidase high-affinity antibody (A8592, Sigma-Aldrich, St. Louis, MO, USA; 1:500 ratio). Tubulin was used as a loading control for protein expression by using an anti-α-tubulin peroxidase-coupled antibody ((11H10)9099, Cell Signalling, Danvers, MA, USA; 1:1000 ratio) incubated overnight. LAMC1 secretion was used as a loading control of protein secretion by using an anti-LAMC1 antibody (HPA001908 (rabbit), Sigma-Aldrich, St. Louis, MO, USA; 1:1000 ratio) incubated overnight, followed by a 1-h incubation with an anti-rabbit peroxidase-coupled antibody (7074, Cell Signalling, Danvers, MA, USA; 1:5000 ratio). Results are representative of three independent experiments.
Immunohistochemistry on lung tissues
Tissue samples from lung biopsies or lung explants were centrally analysed by optic microscopy coupled with computerised image analysis of stained materials. They were also studied by haematoxylin (H) and eosin (E) staining and by immunohistochemistry [28]. A mouse SP-A monoclonal antibody (32E12, Abcam, Cambridge, UK; 1:200 ratio) was used according to manufacturer's protocol. Histopathological descriptions were provided according to the clinical practice guidelines [3] and SP-A staining was analysed on the available tissues.
Results
Spectrum of SFTPA1 and SFTPA2 mutations in 14 independent patients
A total of 11 heterozygous class 4 or class 5 mutations were identified: three SFTPA1 mutations in three probands and eight SFTPA2 mutations in 11 probands (table 1). Eight of them have not been reported so far. All are missense mutations identified in exon 6 that encodes the CRD of SP-A1 and SP-A2. All these mutations involve an invariant or highly conserved residue in SP-A1 and SP-A2 and throughout evolution (figure 1). None of them are prone to alter splicing according to the MaxEntScan tool (http://hollywood.mit.edu/burgelab/maxent/Xmaxentscan_scoreseq.html). Their pathogenicity, as assessed through in silico means, is provided in table 1 [21]. Three SFTPA2 mutations have each been characterised in two distinct families: V178M in families 5 and 6, C238S in families 11 and 12 and R242Q in families 13 and 14 (figure 2). This could be related either to a common ancestor to the two families, or to recurrent mutational events. Families 5 and 6 originate from China and North Africa, respectively and families 13 and 15 from North Africa and La Réunion Island, respectively (table 2). Haplotyping showed that V178M arose on a different haplotype in family 5 (haplotype H2) to family 6 (haplotype H4) (supplementary figure S1a). The R242Q mutation also arose on a distinct haplotype in family 13 (haplotype H11) to that for family 14 (haplotype H14) (supplementary figure S1c). These data, together with the diverse geographic origins of the families and the fact that these two mutations are G-to-A transitions involving a CG dimer, strongly support the hypothesis of recurrent mutational events in discrete populations. Regarding the C238S mutation, haplotype analysis of 17 SNPs did not exclude the existence of an ancestor common to families 11 and 12, which both originated from North Africa (Algeria) (supplementary figure S1b). This variation is not described in the gnomAD database (https://gnomad.broadinstitute.org) and is a G-to-C transversion (mutation mechanism independent of CG dimer hotspots). However, the H6 haplotype characterised in both families is made of the most frequent versions of the 17 SNPs in humans and the occurrence of a recurrent mutational event in a common haplotype cannot be fully rejected.
Structural and electrostatic changes induced by the identified SP-A1 and SP-A2 mutations
As compared to WT SP-A1 and SP-A2 CRDs, prediction of the 3D structure of the protein carrying the mutations disclosed no major consequences on the overall domain structure (RMSD values less than 1.1 between WT and mutated SP-A1 or SP-A2 domains). The electrostatic modelling of the mutant protein domain surface shows polarity changes for six of the mutants, as compared to the corresponding WT SP-A CRD (supplementary figure S2).
Impact of the identified missense variants on SP-A1/SP-A2 production and secretion
As shown on the Western blots performed on whole cell lysates, as well as by the histograms with densitometry ratio from three or more independent experiments (figure 3), the expression of each mutant SP-A1 and SP-A2 protein (35–37 kDa) is similar to the normal proteins in HEK293T lysates (p>0.05, by t-test), thereby indicating that the identified SFTPA1 and SFTPA2 missense variations have no effect on protein expression. However, the expression of each mutant SP-A1 and SP-A2 protein is dramatically reduced in the cell medium (p<0.05, by t-test), thereby demonstrating that these missense variations result in the absence of SP-A1 and SP-A2 secretion and are therefore deleterious.
Parenchymal expression pattern of SP-A in the lung tissue of probands
Tissue samples from lung biopsy or explanted lungs of four probands (figure 4) showed an “indeterminate for usual interstitial pneumonia (UIP)” pattern. The SP-A expression pattern was characterised by marked expression in hyperplastic pneumocytes and discontinuous expression surrounding the alveolar space. Such an expression pattern contrasts with the light expression and thin continuous linear layer of SP-A at the alveolar surface of the control specimen [28]. In all patients, residual SP-A expression was observed in alveolar macrophages, consistent with the fact that the antibody used in these experiments recognised both SP-A1 and SP-A2 (which are 98% identical). The same features were observed on the lung biopsies of two affected siblings (supplementary figure S5).
Clinical characteristics of the 14 patients and their 14 relatives carrying a SFTPA1/SFTPA2 mutation
The detailed phenotypic characteristics of the 14 unrelated probands are provided in table 2. Eight of them (57%) originated from North Africa. Only one proband, aged 29 at ILD onset (family 13, II.1), had neither lung cancer nor family history of lung cancer or ILD. Eleven (78%) had a first-degree history of ILD and nine (64%) had a personal (n=5, 36% assessed by lung biopsy and/or explant analysis) and/or family (n=7, 50%) history of lung cancer (figure 2). A total of 10 proband computed tomography (CT) chest scans could be centrally reviewed by a thoracic radiologist (figure 5) [3]. Only two patients showed a UIP pattern [3]. Half of cases were classified as indeterminate for UIP and three cases suggested either nonspecific interstitial pneumonia (NSIP) or hypersensitivity pneumonitis (HSP). Two patients had apical subpleural consolidations evoking pleuroparenchymal fibroelastosis (PPFE) associated to UIP and indeterminate fibrosis (one case each). In total, predominant ground glass opacities (GGOs) were observed in seven out of 10 patients (70%) and lung fibrosis was observed in seven out of 10 patients (70%). Other inconstant features were basal thickened interlobular septa, micronodules, reticulations, honeycombing and bronchiectasis. The same results were observed on the CT chest scans of two affected siblings (supplementary figure S4).
Family screening led to the identification of an SFTPA1 or SFTPA2 mutation in 14 more individuals (figure 2 and supplementary table S1). Altogether, the present study allowed the identification of 28 individuals with an SFTPA1 or SFTPA2 mutation. The male-to-female ratio was 0.65. Median age at onset of ILD was 45 years (range 0.6–65 years). Five patients (22%) presented an extra-respiratory manifestation, including two patients with rheumatoid arthritis (9%). Ten patients (43%) declared neither tobacco nor professional exposure. Seven relatives remained asymptomatic until a median age of 39 years (range 19–55 years) and after a median length of follow-up of 5 years (range 1–8 years). The penetrance of the disease was incomplete, as two individuals were aged over 40 years (42 years and 55 years, respectively) and harboured normal pulmonary function tests (PFTs) and CT scans. The other asymptomatic relatives had either abnormal PFTs and/or minimal interstitial abnormalities on the CT scan at ages 18, 24, 32, 33 and 39 years, respectively. The first and last available PFTs of the patients are provided in supplementary figure S3. The evolution of the ILD was severe: five patients needed anti-fibrosing therapies and 11 patients (48%) received lung transplantation at a median age of 51 years (range 37–58 years) with a median time from diagnosis of 5 years (range 0–19 years). Regarding lung transplantation, one patient with ILD and surgically treated lung cancer (family 2, IV.3) benefitted from an atypical indication: despite a minimally invasive adenocarcinoma with a lepidic pattern on the contralateral lung almost 10 years after the first, lung transplantation was decided upon, with a 2-year favourable outcome to date. For all the patients the lung transplantation was bilateral in order to prevent lung fibrosis extension and lung cancer development on the native lung. Nine patients (32%) died at a median age of 51 years (range 0.7–68 years), three of them shortly after lung transplantation (7 days, 6 weeks and 13 months, respectively) (supplementary table S1 and figure 6).
Discussion
The present study describes for the first time a large number of functionally assessed pathogenic SFTPA1 and SFTPA2 mutations in 14 independent families. Three mutations were identified in SFTPA1 and eight in SFTPA2, bringing the total number of reported SFTPA1 and SFTPA2 mutations to five and 14 respectively (supplementary table S2) [11, 14–19]. Interestingly, all the described mutations are located in exon 6 that encodes the protein CRD and all but one (p.Tyr208His) were found in the heterozygous state.
In this study, the pathogenicity of the identified mutations was assessed by in vitro functional studies and ex vivo immunostaining, showing that the secretion of all the mutant proteins was similarly abolished and that the protein may be sequestered in type 2 alveolar epithelial cells (AEC2s). Although part of these results relies on a cell model that over expresses recombinant proteins, such that quite unequivocal presentation of the disease at the cellular level contrasts sharply with highly variable clinical expression (which ranges from incomplete penetrance to severe forms requiring lung transplantation).
Phenotypic heterogeneity
The individuals with SFTPA1 or SFTPA2 mutations presented with a heterogeneous clinical phenotype, with no evidence of phenotypic differences between patients with SFTPA1 mutations and those with SFTPA2 mutations. The penetrance of the disease was incomplete, with the identification of carriers remaining asymptomatic in adulthood (normal CT chest scans and PFTs). The age of onset was variable, from childhood to old age [19]; however, in the majority of the patients, ILD occurred at a younger age (mean 40±14 years (range 0.6–65 years)) than the usual age for IPF onset, as only one patient experienced disease onset after 60 years of age. CT-scan analysis gave highly heterogeneous results in terms of the American Thoracic Society (ATS)/European Respiratory Society (ERS) classification, highlighting that this classification is maybe not suitable in such cases. As shown in the current study and as observed in children with surfactant disorders, the GGO pattern represents the most frequent pattern [29]. This observation differs from the typical UIP pattern that is seen in IPF and could be one of the main arguments, on top of the family form of ILD, for suspecting a surfactant disorder in adults. Interestingly, no relationship could be established between the severity of the CT images and the mutations. Moreover, various CT patterns were observed within the same family. Altogether, as is also the case in other causes of genetic ILD, variability in disease expression was observed among different individuals, including among those from the same family [22].
Variability in clinical expression of disease phenotype is a common feature of most genetic disorders, especially when dominant. In theory, this phenomenon may result from allelic heterogeneity and/or from the influence of environmental or modifying genetic factors [30]. Moreover, post-translational modifications of SP-A proteins could at least partly account for the heterogeneous presentation of the clinical phenotype. The role of viruses has been discussed in paediatric patients who develop fatal pulmonary fibrosis [6]. This hypothesis has also been discussed for mutations in other surfactant genes, especially SFTPC and ABCA3 mutations [31, 32]. In adults, tobacco smoking and occupational exposures are supposed to be the main triggers of IPF [6, 33], but the role of viral infections as a risk factor for pulmonary fibrosis has also been reported by several authors and recently confirmed in a meta-analysis [34]. The interaction between SP-A and various viruses has been studied in vitro and in SFTPA-knockout mouse models. The viral infections appeared to promote lung fibrosis development, whereas an adjunct SP-A treatment allowed a beneficial effect on lung fibrosis progression [35]. SP-A binds the pathogens’ oligosaccharides via its CRD, induces their opsonisation and/or phagocytosis and promotes a pro-inflammatory response [36]. Alternatively, the CRD binding to the alveolar macrophages could induce an anti-inflammatory response by enhancing apoptotic cell phagocytosis. This CRD ability to regulate pro-inflammatory and anti-inflammatory responses depending on the orientation of SP-A binding to alveolar macrophages, via a CRD-dependent or CRD-independent mechanism, is known as the “head or tail hypothesis” [37, 38]. In this context, molecular alterations to the SP-A1/SP-A2 CRD could alter its ability to respond to pathogen exposures and to modulate pulmonary inflammation caused by environmental triggers.
The observation of SP-A accumulation in AEC2s in lung tissue sections suggests the production of misfolded proteins in the AEC2s that could induce variable levels of endoplasmic reticulum (ER) stress and promote cell death or cell reprogramming [39]. The unsecreted abnormal SP-A protein could thus be responsible for an unfolded protein response (UPR) challenge. UPR and ER stress have been found to be increased in the case of SFTPC mutations [39]. Evidence for increased ER stress as well as abnormal oligomerisation and increased proteasome degradation has also been reported for mutant SP-A2 proteins (G231V and F198S mutations) [40]. Similarly, SFTPA1 mutations have been associated with an increased expression of ER stress and necroptosis markers (Y208H mutation) [41]. Thus, there is compelling evidence of a major role for ER stress in SFTPA1 and SFTPA2 mutations that needs to be further confirmed in patients’ lung tissues.
Lung cancer association
IPF has been associated with an increased risk of lung cancer, such an association being observed in up to 13.5% of patients with pulmonary fibrosis [42, 43]. In the present study, the association of ILD and lung cancer reaches 36% in the probands and 64% when including family history, which strongly supports a link between SFTPA1 and SFTPA2 CRD mutations and lung cancer development. Interestingly, such associations have not been reported for SFTPC mutations and, given the severity of SFTPC-associated disease leading to lung transplantation or to death at a young age, the follow-up period may not be long enough to allow for tumorigenesis promotion. ER stress via UPR is also linked to tumorigenesis [44]. Increased ER stress can alter tissue homeostasis by interfering with cell cycle in many ways, including cell death or cell senescence attenuation (and thus tumorigenesis promotion) [45]. However, in the case of SP-A mutations, these hypotheses still need to be further evaluated. It has also been proposed that increased expression of SP-A could reduce adenocarcinogenesis in mice [46]. This can be explained by the interaction between SP-A and transforming growth factor-β (TGF-β), a crucial molecule implicated in lung repair and tumour proliferation. SP-A has been shown to inhibit TGF-β1 inactivation [47, 48] and thus mutations of SP-A that alter its secretion could prevent SP-A from inhibiting the TGF-β pathway, leading to an impairment of the alveolar healing process [49].
Genetic studies, family screening and genetic counselling
This study suggests that genetic screening for SFTPA1 and SFTPA2 mutations should be proposed to patients with 1) a family form of ILD, especially if associated with a personal or family history of lung cancer; and 2) an early-onset ILD (before 50 years of age) and no telomerase complex gene mutations. In line with incomplete penetrance, as well as risk of ILD and lung cancer, when a mutation is found a family screening should be offered to the patient's siblings and other relatives, including asymptomatic major relatives. The pre-symptomatic management of individuals with SFTPA1 and SFTPA2 mutations is not yet the subject of consensus, but it is likely that tobacco smoking and environmental exposures can enhance disease expression and that genetic counselling should also include targeted preventive measures. In France, national multidisciplinary team (MDT) meetings dedicated to genetic forms of pulmonary fibrosis have recently been launched in the frame of the RespiFIL network [22]. Among other issues, prenatal and pre-implantation management are discussed for each family. Following the MDT report, a personalised genetic counselling is given to the patient and his family. As shown in this study, the phenotypic heterogeneity of the disease is important and could change the perceptions of the family regarding disease severity and their expectations in terms of prenatal/pre-implantation management.
Conclusions
Heterozygous SFTPA1 and SFTPA2 mutations involving exon 6, which encodes the CRD of the corresponding proteins, contribute to various fibrotic ILDs that can occur in young adults (probably during childhood) and to a higher risk of lung cancer. Their pathogenicity is attested to by impaired protein secretion leading to cytoplasmic retention in the alveolar epithelium. However, the penetrance of the disease phenotype is incomplete and clinical expression of the disease is highly variable, with a major proportion of patients (including young adults) presenting a severe disease requiring lung transplantation. As such, family screening and genetic counselling, as well as pre-symptomatic management of the relatives, are major issues that need to be discussed in the frame of MDT meetings.
Supplementary material
Supplementary Material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary tables ERJ-02806-2020.Tables
Supplementary figures ERJ-02806-2020.Figures
Shareable PDF
Supplementary Material
This one-page PDF can be shared freely online.
Shareable PDF ERJ-02806-2020.Shareable
Acknowledgements
We wish to thank the patients and their families for their participation in the study. We also thank the Assistance Publique Hôpitaux de Paris (APHP), Sorbonne Université (Paris, France) and the national networks for rare lung diseases (the Centre de référence des maladies respiratoires rares (RespiRare), the Centre de référence des maladies pulmonaires rares (OphaLung) and the Filière de soins pour les maladies respiratoires rares (RespiFIL)). The ILD cohort is developed in collaboration with the Rare Disease Cohorts (RaDiCo)-ILD project (ANR-10-COHO-0003), the FP7-305653-child-EU project and the COST Action European Network for Translational Research in Children's and Adult ILD (COST-ILD) project (CA16125).
Footnotes
This article has an editorial commentary: https://doi.org/10.1183/13993003.03252-2020
This article has supplementary material available from erj.ersjournals.com
Author contributions: N. Nathan, M. Legendre, A. Clement and S. Amselem designed the study; N. Nathan, A. Butt, E. Filhol-Blin, T. Desroziers, M. Héry, V. Nau and A. Coulomb L'Hermine performed the experiments; N. Nathan, M. Legendre, B. Copin, P. Duquesnoy, E. Filhol-Blin, T. Desroziers, M. Héry, A. Butt, A. Coulomb L'Hermine, M-P. Debray, A. Clement and S. Amselem analysed and interpreted the data; N. Nathan and A. Butt wrote the manuscript; M. Legendre, S. Amselem, A. Clement, B. Copin, M-P. Debray, A. Coulomb L'Hermine, R. Borie, F. Dastot Le Moal, C. Kannengiesser, L. Gouya, V. Cottin, J. Traclet, B. Crestani, A. Gondouin, C. Dupin, C. Dombret, A. Cazes, A-L. Chene, V. Giraud, H. Nunes, D. Bouvry, A. Bergeron, G. Lorillon, D. Israël-Biet, P. de Vuyst, C. Picard, E. Longchampt, M. Reynaud-Gaubert, J-C. Dalphin, S. Leroy, A. Le Borgne and N. Allou provided the patient data; all authors reviewed and approved the manuscript.
Conflict of interest: M. Legendre has nothing to disclose.
Conflict of interest: A. Butt has nothing to disclose.
Conflict of interest: R. Borie reports grants and personal fees from Boehringer Ingelheim and Roche, and personal fees from Savapharma, outside the submitted work.
Conflict of interest: M-P. Debray has nothing to disclose.
Conflict of interest: D. Bouvry has nothing to disclose.
Conflict of interest: E. Filhol-Blin has nothing to disclose.
Conflict of interest: T. Desroziers has nothing to disclose.
Conflict of interest: V. Nau has nothing to disclose.
Conflict of interest: B. Copin has nothing to disclose.
Conflict of interest: F. Dastot Le Moal has nothing to disclose.
Conflict of interest: M. Héry has nothing to disclose.
Conflict of interest: P. Duquesnoy has nothing to disclose.
Conflict of interest: N. Allou has nothing to disclose.
Conflict of interest: A. Bergeron has nothing to disclose.
Conflict of interest: J. Bermudez has nothing to disclose.
Conflict of interest: A. Cazes has been invited to national and international meetings, and/or has received grants and/or personal fees for various activities from Boehringer Ingelheim and Roche, outside the submitted work.
Conflict of interest: A-L. Chene has nothing to disclose.
Conflict of interest: V. Cottin reports personal fees for lectures and advisory board work and non-financial support for meeting attendance from Actelion; grants, personal fees for lectures and advisory board work, and non-financial support for meeting attendance from Boehringer Ingelheim; personal fees for advisory board and data monitoring committee work from Bayer/MSD and Galapagos; personal fees for lectures and advisory board work from Novartis; personal fees for lectures, consultancy, data monitoring committee and steering committee work, and non-financial support for meeting attendance from Roche/Promedior; personal fees for lectures from Sanofi and AstraZeneca; personal fees for data monitoring committee work from Celgene and Galecto; and personal fees for advisory board work from Shionogi, outside the submitted work.
Conflict of interest: B. Crestani has nothing to disclose.
Conflict of interest: J-C. Dalphin has nothing to disclose.
Conflict of interest: C. Dombret has nothing to disclose.
Conflict of interest: B. Doray has nothing to disclose.
Conflict of interest: C. Dupin reports personal fees for lectures and advisory board work, and non-financial support and meeting invitations from AstraZeneca; personal fees for lectures, non-financial support and meeting invitations from Boehringer and Novartis; personal fees for research and lectures, and non-financial support and meeting invitations from GSK; personal fees for lectures and meeting invitations from Chiesi; personal fees for lectures and advisory board work, and non-financial support from Sanofi; and personal fees, non-financial support and meeting invitations from Roche, outside the submitted work.
Conflict of interest: V. Giraud has nothing to disclose.
Conflict of interest: A. Gondouin has nothing to disclose.
Conflict of interest: L. Gouya has nothing to disclose.
Conflict of interest: D. Israël-Biet has nothing to disclose.
Conflict of interest: C. Kannengiesser has nothing to disclose.
Conflict of interest: A. Le Borgne has nothing to disclose.
Conflict of interest: S. Leroy has nothing to disclose.
Conflict of interest: E. Longchampt has nothing to disclose.
Conflict of interest: G. Lorillon has nothing to disclose.
Conflict of interest: H. Nunes has nothing to disclose.
Conflict of interest: C. Picard has nothing to disclose.
Conflict of interest: M. Reynaud-Gaubert has nothing to disclose.
Conflict of interest: J. Traclet has nothing to disclose.
Conflict of interest: P. de Vuyst has nothing to disclose.
Conflict of interest: A. Coulomb L'Hermine has nothing to disclose.
Conflict of interest: A. Clement has nothing to disclose.
Conflict of interest: S. Amselem has nothing to disclose.
Conflict of interest: N. Nathan reports a 2018 AstraZeneca Mobility Grant from Société de pneumologie pédiatrique et d'allergologie (France), outside the submitted work.
Support statement: This work was supported by grants from the Institut National de la Santé et la Recherche Médicale (INSERM), the Legs Poix from the Chancellerie des Universités de Paris (grants 2013 number 1305, 2014 number 1405, 2015 number 1015, 2016 number 2077 and 2017 number DP2017/1860), the Société Française de Pédiatrie - Société Pédiatrique de Pneumologie et d'Allergologie and AstraZeneca, as well as funding from the patient organisationsRespirer c'est Grandir and the Belleherbe Association. Funding information for this article has been deposited with the Crossref Funder Registry.
- Received April 26, 2020.
- Accepted July 22, 2020.
- Copyright ©ERS 2020