Outcome of lung transplantation for adults with interstitial lung disease associated with genetic disorders of the surfactant system

Discussion

In this retrospective series, we studied the clinical outcome of adult patients who underwent LTx for ILD associated with GDS. With a median overall survival of 8.6 years, we demonstrate that LTx is a therapeutic valid option for these patients.

To our knowledge, this is the first assessment of lung transplanted adults in such context of rare genetic disorders including exclusively pathogenic or probably pathogenic mutations. However, this study meets limitations; indeed, its retrospective design and the few number of patients included makes necessary to interpret these results with caution.

This is an extremely rare situation given that only 20 adult recipients were identified over >4000 LTx in France from 1997 to 2018 (Agence de la Biomédecine, France). Regarding to LTx for TRG mutation, 38 LTx were performed in France in the period from 2008 to 2018 [15], highlighting how rare GDS are. However, it is likely that patients with such genetic disorders were underdiagnosed, at least in the first 10 years of the study, when the systematic genetic testing for family forms of ILD was far from exhaustive. Indeed, for a long time, research for GDS in ILD has been carried out exclusively in infants and children and has been offered only recently to young adults with a history of ILD [18].

Very few data are available in lung transplanted patients with ILD associated with GDS; to date, it has only been reported in a selected paediatric population in the North America. Herein, mostly ILDs related to SFTPA1 and SFTPA2 were included, consistent with the fact that these surfactant disorders manifest mostly, but not exclusively, in adulthood. An increasing number of ILDs related to SFTPC [19] and more rarely ABCA3 [20] mutations in adult patients are also being reported, either in patients diagnosed in childhood who reached adulthood, or in patients who developed symptoms in adulthood. In this study, eight patients had a SFTPC mutation, but interestingly, none had ABCA3 mutation. These patients were excluded because they had mutation of uncertain significance, although they were treated as if they had a pathogenic/likely pathogenic variant. In addition, patient with non-null homozygous ABCA3 mutations may show an improvement with immunosuppressive treatments, as recently reported [21, 22], and may eventually show better prognosis and less LTx. These two points highlight the role of expert genetic opinion and dedicated multidisciplinary discussion for their evaluation.

Median age at onset of respiratory symptoms was 42 years and 51 years at LTx, but one-third showed a sudden deterioration requiring an emergency LTx. These results suggest that despite their young age, patients should be referred to LTx centres at diagnosis to reduce the rate of high-emergency LTx and eventually improve the post-transplant survival. A young age at diagnosis and LTx has also been reported in patients with TRG mutations with symptoms beginning at ∼51 years and LTx ∼61 years [23]. Our data suggest that patients with ILD associated with GDS may develop symptoms even earlier than those with TRG mutations, and will need earlier LTx, highlighting the need to diagnose GDS in early forms of ILD before 50 years.

Median survival in our cohort was 8.6 years, which is better than that seen in patients transplanted for idiopathic and non-idiopathic ILD, for whom median survival was 5.2 years and 6.7 years, respectively, in 2019 [7]. This difference may be explained by some different demographic characteristics before LTx, with a younger age and fewer comorbidities. However, these data should be interpreted with caution given the small number of patients included. Even if our median survival was acceptable, three patients died within 3 months because of surgery complications. This data can be explained by the fact that 30% of patients were transplanted in high-emergency procedures, which are associated with a higher post-transplant mortality. Moreover, patients with ILD associated with TRG mutations will be at risk of developing other life-threatening disorders such as myelodysplasia, bone marrow failure and liver disease. These may not only develop before LTx, but also post-transplant, possibly triggered by immunosuppressive treatments or CMV infection, which will worsen prognosis [24]. In our study, few patients developed haematological complications (anaemia was secondary to renal insufficiency or iron deficiency) after LTx compared with patients with TRG mutations. None of them developed any myelodysplastic syndrome or malignant haemopathy, and CMV mismatch was not associated with a worst prognosis. These data would prompt the use of immunosuppressive and potentially haematotoxic drugs after LTx according to the usual protocol of each centre, unlike for TRG mutation carriers.

GDS, and especially SFTPA1 and SFTPA2 mutations, may be associated with lung cancer [25, 26] which is a classical contraindication to LTx within 5 years after cancer treatment [27]. However, in our series, two patients with an SFTPA1 mutation had a history of localised lung cancer in the 5 years preceding LTx. They did not present a recurrence of lung cancer during follow-up and were still alive at the end of the study, with follow-up of 2.1 and 7.4 years. Although preliminary, these data could be a starting point to discuss lung cancer as a usual contraindication for LTx in candidates with GDS [28]. Conversely, two other patients developed lung cancer after LTx. One of them had an undiagnosed lung cancer at LTx, and the other one developed cancer a few years later after unilateral LTx. In a series of adult patients with ILD associated with GDS, a patient with a SFTPA2 mutation, who underwent an initial single left LTx, had been retransplanted with bilateral LTx due to the occurrence of CLAD. This patient experienced a lung adenocarcinoma (diagnosed on the native right explanted lung) which led to death a few months following the retransplant [5]. These results emphasise the importance of detecting lung cancer in these young at-risk patients before LTx by close monitoring. Furthermore, in such patients, the potentially fast worsening of respiratory function, and the poor prognosis without LTx, could shorten the delay of 5 years classically expected before LTx under cover of close cancer screening. Finally, a systematic bilateral LTx should be a better option to minimise the risk of cancer on the native lung in this at-risk population.

In post-transplant period, two of our 20 patients died of lung cancer. Even if our population is too small to make conclusions, this rate is higher than that found in recent published data showing an incidence of lung cancer in lung transplant recipients of 528 per 100 000 person-years [29]. In this way, these patients should be more closely screened for lung cancer especially patients with unilateral LTx with at least an annual computed tomography (CT) scan.

In our study, survival of lung transplant recipients with a SFTPA1 mutation was found significantly better compared to patients with other genes involved, whereas the demographics were similar regarding age at diagnosis and ILD histological pattern. However, the SFTPA1 mutation group consisted mainly of women, whereas there were mostly men in other groups of mutations; it is understood that registries of LTx such as the one kept by the International Society for Heart and Lung Transplantation shows that transplanted women had better prognoses than men [7]. Another hypothesis may be that pathogenicity of surfactant mutation is related to their structural and immune-modulatory function, which is essential for tissue homeostasis and to avoid chronic inflammation that promotes fibrosis. Given that the function of these proteins are different, it can explain that these mutations may have a different impact on the prognosis after lung transplant. However functional assessment of these mutations is currently limited and insufficient to find out a real mechanism explaining this difference [4]. These hypotheses will have to be confirmed in a larger group of patients allowing a multivariate analysis.