Screening for Preclinical Pulmonary Fibrosis: Are We There Yet?

Progressive pulmonary fibrosis (PPF), although not a specific diagnosis, is associated with progressive dyspnea, worsening lung function and radiographic changes, reduced quality of life, and mortality (1, 2). Antifibrotic drugs are associated with decreased disease progression across a variety of progressive fibrotic lung diseases, generating enthusiasm to identify individuals at risk of disease development and progression at an early stage (3).

Individuals with two or more first-degree relatives (FDRs) with pulmonary fibrosis (PF) are at increased risk for interstitial lung abnormalities (ILAs) and interstitial lung disease (ILD), although studies have also shown increased risk in FDRs of sporadic PF patients (4). Among patients with idiopathic pulmonary fibrosis (IPF) and non-IPF ILDs, patient-provided family history of ILD alone is associated with increased ILD risk and reduced transplant-free survival compared with those with sporadic disease (5). A recent 2-year follow-up study of the FDRs of PF patients found that 20/53 had ILAs at baseline, with radiographic progression in 65% of those with ILA (6). However, a Fleischner Society ILA position paper noted radiographic abnormalities on computed tomography (CT) chest screening of high-risk groups (including those with familial ILD) would not be considered ILAs and would be better referred to as preclinical ILD (7).

In this issue of the Journal, Steele and colleagues (p. 587–593) present longitudinal data from a large prospective cohort of FDRs >40 years old from families with two or more patients with familial interstitial pneumonia (FIP) with a goal of defining incidence, progression, and features of what the group previously described as preclinical PF (prePF) (8, 9). High-resolution CT (HRCT) of prePF patients showed reticulation with equivocal or present traction bronchiectasis with or without honeycombing. Among 493 asymptomatic FDRs, 15.5% had prePF at baseline. Sixty percent of the cohort completed follow-up health questionnaires and repeat CT chest screenings. Over a median follow-up of 3.9 years, 6.3% of FDRs without prePF on initial screening developed prePF, translating to an annual incidence of 1.6%. This is 100-fold to 150-fold higher than the incidence of IPF in the general population. The authors found that prePF at baseline was associated with decreased survival: 11.7% of prePF patients died (6/9 respiratory related) compared with 2.2% without prePF (0/9 respiratory related).

This study has several strengths. It is the largest prospective study of FDRs from FIP families. The prePF designation was made by both blinded radiology core visual interpretation of HRCT and clinical core consensus among members who independently reviewed radiologic and clinical data. Phenotyping was robust, with adjudication by a third radiologist or interactive consensus in the radiologic and clinical cores, respectively, if there was disagreement. There were nine individuals who had prePF at baseline but not on follow-up. The authors provide detailed review and description of these subjects. Ultimately, initial radiographic abnormalities in these patients were subtle with interobserver variation. Fibrosis score by quantitative HRCT using data-driven textural analysis, associated with reduced survival, was noted to potentially be helpful in such cases. The authors provide follow-up on subjects with initial CT indeterminate for fibrosis: 4/15 developed prePF. This highlights that FDRs without fibrotic changes on initial CT may develop fibrosis over time. Genetic testing was included. Minor allele frequency of the gain-of-function MUC5B promoter variant rs35705950 was significantly increased in prevalent prePF at baseline but not at follow-up. The MUC5B promoter variant was not significantly associated with survival; the authors rightly note that sample size limited the assessment of genetic risk factors.

The authors conclude that HRCT screening should be considered for FDRs over age 50 from FIP families. Given the known high costs of IPF and PPF in symptoms, loss of life, and healthcare spending, it seems prudent to determine how to best screen those known to be at highest risk (10). A primary question arising from this study is optimal mode of screening, monitoring, and time frame. Screening by family history is appealing, given its low cost, but would potentially miss disease in individuals for whom family history is unknown or where families are small. Up to 25% of patients with FIP may have mutations in telomere-related genes, making them more susceptible to the adverse effects of radiation, raising concerns about serial HRCT (11, 12). Although quantitative CT with data-driven textural analysis may be particularly helpful in detecting subtle abnormalities that track with mortality, it is neither widely available nor standardized. Genetic and serum biomarkers are promising, although genetic variants remain unknown in up to 75% of familial PF kindreds such that negative predictive value of such testing at present would be insufficient (11). Further study of screening modalities and optimal timing for initiation and monitoring is needed.

Limitations of the study include difficulty in discerning who among the initially asymptomatic prePF patients develop clinical PF. The clinical core assigned diagnoses, including IPF, although pulmonary function testing (PFT) and 6-minute walk data were not collected. Subjects completed dyspnea questionnaires but not a detailed clinical history, an essential component of PF diagnosis (13). Although a significantly greater proportion of patients with prePF had increased dyspnea scores over 4 years compared with those without prePF, the presence or degree of physiologic impairment is not known. The addition of PFT data may have helped distinguish prePF and PF, although the utility of adding PFT data to predictive models has been mixed in studies seeking to distinguish affected from healthy FDRs from FIP families (4, 14, 15). The study cohort consisted of non-Hispanic White patients. Although this is often the case in studies of PF, it is a significant limitation. Future studies should be enriched to include other ethnic groups, particularly considering known racial disparities in PF diagnosis, treatment, and outcomes (16, 17).

Hopefully, the current study by Steele and colleagues will provide impetus for future prospective studies of screening for prePF and whether early treatment may affect outcomes. Some programs provide screening, which this is best done through a multidisciplinary team to help patients navigate results. There are currently no guideline recommendations surrounding screening, and this often limits insurance coverage for such testing in the United States. Another important question is whether prePF should be considered a diagnosis, similar to prediabetes. Further consensus and guideline recommendations to streamline modality and time frame of screening as well as terminology will likely be needed before this is feasible. Although we may not be quite there yet from the perspective of screening in routine clinical practice, this study by Steele and colleagues is an important step in closing the gap.