The PICO (Population-Intervention-Comparator-Outcome) question format was employed to guide the literature search, posing the following inquiries: What are the best procedures for the diagnosis of lung diseases in patients with NKX2-1-related disorders? and What are the best procedures for treatment and follow-up of lung diseases in patients with NKX2-1-related disorders? (Supplementary Data 1). These questions covered all the aspects related to the detection, diagnosis, treatment, and follow-up of respiratory manifestations in NKX2-1-RD. The protocols of the systematic reviews for each question were previously registered in the International Prospective Register of Systematic Reviews (PROSPERO) repository with the identifications CRD42023448437 and CRD42023448440.

Specific inclusion criteria were employed to select the relevant studies for these systematic reviews. The population criteria were common for both research questions: patients of all ages with genetic confirmation of the disease (variants in NKX2-1, previously known as TTF-1, or deletion in the 14q13.3 chromosome) with respiratory manifestations. Non-human studies and patients without genetic confirmation of the disease were excluded. Given the rare condition, any comparator was used.

The following specifications apply to the interventions and their outcomes: For diagnosis, the interventions included physical examination, pulse oximetry test, blood tests (blood gases, hemogram, reactive C protein, blood culture, etc.), pulmonary function testing (spirometry, bodyplethysmography), exercise tolerance testing, diffusion capacity for carbon monoxide (DLCO), imaging (chest X-ray, CT scan, PET-CT scan), lung biopsy (transbronchial forceps- or cryobiopsy, open- or thoracoscopic biopsy), and outcomes were determined based on test results related to the diagnosis. Regarding treatment, the interventions comprised supplementary oxygen, noninvasive and invasive respiratory support, medication (bronchodilators, corticosteroids, immunosuppressive therapy, antibiotics, exogenous surfactant, etc.), chest physiotherapy, lung transplant, cancer surgery, radiotherapy, chemotherapy, and supportive care, including nutrition, immunization, and environmental care. Outcomes were assessed in terms of the indication for hospitalization, treatment effectiveness, adverse effects of treatments, treatment in special situations (pregnancy), and follow-up (lung outcomes, patient and professional education, quality of life, and self-management in adulthood).

The included study designs were primary studies, systematic reviews, and randomized controlled trials (RCTs). Narrative reviews, protocols, conference articles, editorials, letters to the editor, and those whose full text was unavailable to be retrieved were excluded. While the inclusion criteria allowed articles in all languages, all retrieved articles were written in English.

A systematic literature search was carried out for each research question to review the scientific evidence. MEDLINE (Ovid), Embase (Elsevier), the Cochrane Library (Wiley), APA PsycInfo (EBSCO), CINAHL Complete (EBSCO), the Web of Science (WOS), the TRIP PRO medical database, and the International Health Technology Assessment (INAHTA) database were searched as reference databases, and additional databases were included to retrieve gray literature such as the Cochrane Controlled Register of Trials (CENTRAL) and subject-specific information resources on rare diseases, Orphanet, EURORDIS, NORD, RARE-Best Practices, and Gene Reviews.

The searches covered the period from January 2002 to August 2022, and were updated in July 2023. The included records date back to 2002, when the NKX2-1 gene variants were first discovered and named. Both controlled language (descriptors) and free terminology were used to search for studies. The initial strategies were carried out in MEDLINE (Ovid) and later adapted to each database’s syntax. The Medline searches for each systematic review are available in Supplementary Data 8.

The identified references were imported into the reference management section of the Covidence Systematic Review Software (Veritas Health Innovation; 2023, https://www.covidence.org/), and the duplicates were automatically removed. A different review was created for each question. Two authors independently filtered the references by title, abstract, and full text (KM, ARR) according to the selection criteria. The disagreements between the reviewers were resolved by a third researcher (JDOE). The data were extracted by two independent authors (KM, ARR) and recorded in Excel spreadsheets. Detailed information from the studies, including authorship, publication year, study design, and location, was systematically extracted. Furthermore, specific clinical data for each patient, encompassing age, gender, results of genetic tests, pulmonary diagnosis, diagnostic procedures, phenotype, treatment modalities, follow-up information, and other relevant data, were meticulously documented.

The methodological quality assessment of each study was conducted independently by two authors (KM, ARR), and discrepancies in individual assessments were resolved through discussion to achieve consensus. Specific tools tailored to each study type were employed for quality assessment. The identified references were categorized as case reports, case series, and cohort studies. Case reports and case series were assessed using the tool developed by Murad et al. (13) that evaluates methodological quality in low-evidence studies crucial for decision-making. The tool consists of eight items across four domains (selection, ascertainment, causality, and reporting). Article quality was rated based on the sum of binary responses (1 = low risk of bias, 0 = high risk of bias) for each item, resulting in a score ranging from 0 (poor quality) to 8 (high quality) (14). Adaptations to the ascertainment, causality, and reporting items were made to suit the specific clinical scenario addressed in the research questions.

During the initial search, a total of 4,576 records were located. After removing duplicates, 4,560 potentially relevant studies remained. Title and abstract screening were carried out, yielding 131 studies. Nine references were added after specialist consultants (15–24). While conducting additional NKX2-1 reviews addressing different questions than those posed by the respiratory PICO, the authors observed that the systematic search did not retrieve these nine studies. These studies were included later due to their focus on respiratory symptoms of the disease, which, although important, were not the primary target of the initial search strategy. These studies primarily addressed other affected systems, and it is likely that they were not indexed under the search terms we used, which is why they were not retrieved in the initial search. After further evaluation, it was determined that these studies included patients with respiratory symptoms, justifying their inclusion.

According to inclusion and exclusion criteria, 51 studies were included following full-text screening. 13 included studies were shared by the two PICO questions. Therefore, 38 studies were finally analyzed (15–23, 25–53). Additionally, three references (24, 54, 55) were initially included in the PICO questions. However, during data extraction, the reviewers identified that these studies referred to patients already included in other studies. The two independent screening processes (PRISMA flowcharts) are depicted in Figures 1A,B. The list of included and excluded articles after full-text review, along with the reasons for their exclusion, is shown in Supplementary Data 2. The excluded studies were classified according to the first criterion that did not fit the PICO format.

44.7% of the included studies were case series (n = 17), and 55.3% were case reports (n = 21). No cohort studies were identified. Due to the rare clinical condition studied, no anticipated randomized controlled trials or systematic reviews were available. 44.7% of the data originated from Europe (n = 17), 44.7% from America (n = 17), 7.9% from Asia (n = 3), and 2.6% from Australia (n = 1) (Supplementary Data 3).

The data were collected from 148 patients, comprising 56 males and 67 females, and 25 individuals with unknown sex (Figure 2A; Supplementary Data 4). Among the 92 cases with available gestational age data, 82 were classified as term and 10 as preterm (Figure 2B). Newborn screening for hypothyroidism was available in 115 out of 148 patients, with 20 testing positive and 95 negative. Respiratory symptoms typically manifest at a mean age of 4.4 years, with a median of 30 days (neonatal period), spanning from birth to 73 years.

The initial respiratory symptoms were diverse, including neonatal respiratory distress syndrome (RDS) in 70 patients, asthma and/or wheezing in 20 patients, ILD (with or without hypoxemia), including pulmonary fibrosis in nine patients, severe or recurrent infection with chronic cough in 7 patients, tachypnea and/or hypoxemia in 13 patients, and lung cancer in 1 patient.

In relation to the NKX2-1-RD phenotype, diverse clinical presentations were observed within the studied cohort, both at diagnosis and throughout the disease course. All 148 patients had pulmonary involvement; this was exclusive in only 15 patients, while five showed pulmonary and endocrinological manifestations. Additionally, 19 patients displayed a combination of pulmonary and neurological involvement. The most prevalent category encompassed patients with pulmonary, endocrinological, and neurological manifestations, totaling 86 individuals with this set of phenotypes. 23 subjects had incomplete information that prevented a precise determination of their phenotype, and 4 additional patients were excluded due to issues with the available data. These patients were not included in the detailed analysis. Similarly, in the following, we use only the informative cases available as the denominator.

NKX2-1-RD demonstrates phenotypic respiratory diversity along with endocrinological and neurological involvement. Among the patients, neonatal RDS not progressing to ILD was observed in 52 out of 143 cases (male/female 17:25, unknown 10), while neonatal RDS progressing to ILD was noted in 30 out of 140 patients (male/female 19:9, unknown 2). The difference in sex distribution between these groups was statistically significant (χ² = 4.01, p = 0.045; Fisher’s exact test p = 0.030), suggesting a potential association between male sex and the progression to ILD. Recurrent infections were reported in 59 out of 99 cases (male/female 15:23, unknown 21), and ILD without neonatal RDS was present in 27 out of 142 patients (male/female 8:19). Chronic respiratory insufficiency occurred in 38 out of 123 cases (male/female 19:19), pulmonary hypertension in 15 out of 135 patients (male/female 7:8), and asthma, recurrent bronchial obstruction, or wheeze in 27 out of 133 cases (male/female 12:11, unknown 4). Additionally, lung cancer was observed in three out of 115 patients (male/female, 1:2) (Figure 3). The alluvial diagram and the multi-step alluvial diagram in Figures 4A, B illustrate the interplay of distinct phenotypes in patients exhibiting respiratory manifestations and variants or deletions in NKX2-1.

Endocrinological involvement was present in 90 out of 131 cases, and neurological involvement was reported in a substantial 104 out of 124 cases.

Among the 15 cases with possibly exclusive respiratory manifestations, the distribution of respiratory phenotypes is presented in Table 1.

For a very small fraction of patients, detailed respiratory information (including imaging) was available. Chest radiography (CR) was available for 10 out of 148 patients, with six showing diffuse bilateral opacities and three displaying bilateral opacities alongside other abnormalities such as infiltrates, pneumothorax, or atelectasis, and one case presented with parahilar infiltrates.

CT scans were reported for 34 out of 148 patients; focal consolidations were identified in 12 cases; and ground glass opacities were evident in 24 cases. Additionally, septal or pleural thickening, cystic lesions, bronchiectasis, and architectural distortion were each observed in four cases. Features consistent with usual interstitial pneumonia and a fibrosing interstitial pattern were present in three cases each. Furthermore, two cases exhibited nodular opacities, hilar adenopathy, and paraseptal emphysema.

Regarding pulmonary function test results, nine patients had detailed outcomes (Supplementary Data 5). All patients were older than 6 years (mean 32.2 years, median 25 years, range 8–61 years). Among the patients with results, the pulmonary phenotype included interstitial lung disease without neonatal RDS in 8 patients, recurrent infections in 4 patients, chronic respiratory insufficiency in 3 patients, neonatal RDS progressing to ILD in 1 patient, and lung cancer in 1 patient.

In addition to the pulmonary function tests, bronchoalveolar lavage (BAL) analyses were reported in 11 out of 148 patients; findings revealed neutrophilic inflammation in five cases, decreased levels of surfactant proteins B (SP-B) and C (SP-C) in two cases, and lymphocytic inflammation in one case.

16 histopathological descriptions of lung biopsies were conducted, two of which were in the context of autopsies (at 28 and 40 days of life). The average age, median, and age range of the histopathological study were 14.15 years, 0.75 years, and a range of 23 days to 64 years, respectively. The lung biopsies exhibited a spectrum of histopathological findings, including alveolar growth abnormalities in 9 cases, increased alveolar macrophages in 8 cases, type II cell hyperplasia in 8 cases, alveolar filling with PAS-positive material in 5 cases, alveolar wall thickening and remodeling in 4 cases, pulmonary hypertensive arteriopathy, interstitial pneumonia in 3 cases, foamy macrophages in 2 cases, emphysema, interstitial fibrosis, chronic inflammation, interstitial lymphocytic infiltrate, neuroendocrine cell hyperplasia (via bombesin stain), and non-specific interstitial pneumonia in 1 case each.

In the genetic analysis of patients, a total of 120 variants and 27 deletions in the NKX2-1 gene were reported (Figure 5). The types of variants, listed in order of frequency, included missense (52 cases), nonsense (31 cases), frameshift (17 cases), splicing (9 cases), in-frame insertion (2 cases), microduplication, and null variants (1 case each). Notably, one patient exhibited concurrent in-frame insertion and nonsense variants. Figure 6 illustrates the distribution of NKX2-1 gene variants and deletions categorized by the respiratory clinical phenotype of the patients. Notably, all patients (n = 3) diagnosed with lung cancer exhibit variants classified as nonsense mutations.

Treatment was varied, reflecting the complexity of the respiratory conditions observed. Oxygen supplementation was provided to 76% of patients (66 out of 86), with a mean duration of oxygen therapy (based on 41 patients) lasting 29.4 months. Non-invasive ventilation (NIV) was applied in 8 out of 34 cases (22, 27, 36, 37, 41, 44), while invasive ventilation was employed in 39 out of 70 patients. High-frequency oscillation (HFO) was utilized in 4 out of 48 cases (27, 33, 52, 53), inhaled nitric oxide (iNO) was administered to 5 out of 42 patients (22, 33, 44, 52, 53), and extracorporeal membrane oxygenation (ECMO) was used in 7 out of 82 patients (29, 36, 45, 53).

In terms of pharmaceutical interventions, prescriptions were distributed as follows: Long-term antibiotics were prescribed for 2 out of 33 cases, short-term antibiotics for 16 out of 32 cases, surfactant for 14 out of 47 cases, inhaled corticosteroids (ICS) for 4 out of 21 cases, systemic steroids for 27 out of 46 cases, azithromycin for 18 out of 51 cases, hydroxychloroquine (HCQ) for 10 out of 51 cases, and pirfenidone or nintedanib for 1 out of 29 cases (41).

Age at the last visit was available for 73 out of 148 patients (49%), with a mean of 14.8 years, a median of 6.3 years, and a range from 0.1 to 73.0 years. The duration of follow-up was provided for 52 out of 148 patients (35%), with a mean duration of 5.6 years, a median of 3.3 years, and a range from 0.1 to 24 years. The pulmonary status at the last follow-up revealed diverse outcomes among the 71 patients for whom details were available. These outcomes included 16 patients with no respiratory symptoms, 13 with chronic respiratory insufficiency, 13 who succumbed to respiratory failure, 11 with mild symptoms (sick but improved, dyspnea at rest, chronic cough, exercise intolerance), 6 with recurrent infections, 5 with pulmonary fibrosis, and 4 who passed away due to causes unrelated to respiratory insufficiency (such as lung cancer, pulmonary arterial hypertension, etc.). Five patients underwent double lung transplantation (DLTX), with an age range spanning from 7 months to 13 years (29) (Supplementary Data 6). All these patients presented with chronic respiratory insufficiency: 4 had RDS progressing to ILD, and 1 had ILD without neonatal RDS. Additionally, 3 patients experienced recurrent wheezing, asthma, or obstructive lung disease, while information on respiratory status was not available for 77 patients. The overall survival of informed cases was about 60%.

Concerning the group with recurrent infections, age at the last visit was ascertainable for 34 out of 58 patients (59%), demonstrating a mean age of 12.3 years, a median of 5.5 years, and a range spanning from 4 months to 63 years.

Figure 7A illustrates the distribution of patients across various respiratory phenotypes, stratified by vital status. Figure 7B displays the Kaplan–Meier survival curve of patients with respiratory pathology, comparing those who have undergone lung transplantation (LTX) and those who have not.

According to the assessment tools used in our systematic review, 79% (n = 30) of case series and case reports had poor methodological quality, 18% (n = 7) had medium quality, and 3% (n = 1) had good quality. The scores obtained for each item, as well as the total score for each reference, are listed in Supplementary Data 7.

Table 2 delineates the considerations for suspecting NKX2-1 in patients with various respiratory phenotypes, highlights the distinctions in treatment compared to NKX2-1-negative patients with similar respiratory phenotypes, and provides specific key points for each pathological subgroup.