Defining and Promoting Pediatric Pulmonary Health: Developing Biomarkers for Pulmonary Health

Genetics underlie pulmonary diseases in children, and genetic disorders impact long term pulmonary outcomes. For example, more than 2100 genetic variants have been included in the Clinical and Functional Translation of the cystic fibrosis transmembrane conductance regulator (CFTR) database.¹¹  Of those, more than 300 are thought to be pathogenic.¹²  Like CF, PCD, which might manifest as chronic cough and recurrent sinus, ear, and lung infections, has been associated with more than 50 genes involved in complex pathways for ciliated cell generation and assembly of motile cilia and is more common than previously thought.¹³  This complex genetic variation in PCD‐related genes is associated with distinct clinical phenotypes. Genetics also underlie surfactant production, and symptoms can manifest during the neonatal period and infancy. Although less frequently seen than CF and PCD, childhood interstitial lung diseases related to surfactant metabolism dysfunction has been linked with genes associated with surfactant production (ABCA3, SFTPC, NKX2‐1/TTF1, SFTPB, SFTPA1), surfactant catabolism (CSF2RA, OAS1, CSF2RB, MARS, GATA2), lung development (FOXF1, FLNA, TBX4), and immune dysregulation (TMEM137).¹⁴  Genetic testing results can circumvent invasive diagnostic testing, such as open lung biopsy and inform therapy (precision medicine). Genes that may predict poorer long‐term outcomes for asthma, bronchiolitis, bronchopulmonary dysplasia, and other more common childhood lung conditions could serve as biomarkers to guide personalized therapies. Of note, being heterozygous – even for autosomal recessive conditions‐ may subtly predispose to respiratory disorders that affect long‐term lung health; genetic information has tremendous potential for identifying risk factors for poor long‐term lung health.¹⁵ 

Several knowledge gaps exist regarding the routine use of genetic testing for children with lung diseases. Many lung diseases still have not been well characterized by gene sequencing. Even for those diseases that have been extensively genetically evaluated through sequencing, such as cystic fibrosis, a large number of variants of unknown significance remain.⁴  Because of this, there is a high likelihood of uncovering new variants for which genetic evaluation is not currently helpful.

The prevalence in gene variants differs by ancestry and does not accurately predict disease risk or response to therapy in minority groups. CF occurs in all racial and ethnic populations, though the prevalence of CFTR pathogenic variants vary by ancestry. This has significant clinical implications. For example, approximately 90% of non‐Hispanic white patients with CF have US Food and Drug Administration‐approved genetic variants for CFTR modulators.¹⁶  In contrast, only 69.7% of Black or African American patients and 75.6% of Hispanic patients are eligible for such therapies. These US Food and Drug Administration‐approved CFTR modulator therapies were predominately studied in non‐Hispanic white populations and their frequency and clinical significance in minority population has not been well studied. This has created significant racial and ethnic disparity in the care of heritable diseases, such as CF, where therapy depends on the genetic architecture and the ancestral differences in pathogenic variants. Thus, ancestry specific genetic variant profiles need to be defined in minority populations to determine prognosis and eligibility for therapies guided by genetic variants (eg, CFTR modulator therapy).

Key barriers to addressing these knowledge gaps include cost, a disparity of genetic testing availability between facilities, and silo effects. Not only is the general care of these patients split between general pediatricians and subspecialists, such as allergists and pulmonologists, but complex genetic evaluation often involves geneticists and genetic counselors. For diseases that have not yet been extensively sequenced, this may require establishing new relationships between these silos.

The sequencing of the human genome enabled the discovery of targeted therapies aimed at the biological mechanisms underlying both malignant and nonmalignant diseases. With the wide use of CFTR modulator therapy, the treatment of CF has evolved from treating symptoms and infections to improving the function of a defective cystic fibrosis transmembrane conductance regulator (CFTR) protein and targeting the underlying biologic mechanisms causing CF. This led to improvement in lung function and reduction in the frequency of pulmonary infections and exacerbations. Hence, by treating the underlying biological mechanisms, targeted therapies can improve morbidity and mortality.^17,18  This only emphasizes the importance of early diagnosis and interventions to prevent long‐term damage and increase the lifespan of patients with heritable diseases such as CF.

Despite continually improving evidence‐based guidelines and recommendations for the diagnosis and treatment of pediatric pneumonia, it remains the leading cause of death for children worldwide. Pulse oximetry has become a routine part of the standard of care for management of patients with pneumonia and other respiratory disorders. This technology has become more portable and affordable in the past decade, making it even easier to use in situations of economic and resource restraint. It can also provide extra insight into the respiratory status of pediatric patients who may not be able to communicate effectively with their providers. However, pulse oximetry has limitations.^19–21  Correct application and interpretation of this technology is crucial to provide optimal care to patients.

Most gaps associated with using pulse oximetry in the pediatric setting are related to availability and standardized training.²¹  Although more portable and available than ever, many healthcare settings still lack this technology.⁹  Similarly, training in appropriate use, interpretation, and the limitations of pulse oximetry is not universally available.

Also, more studies are needed to comprehensively assess interdevice reliability, as different devices are used between settings but are compared with the same normal ranges for patients.

Assessing device accuracy without validation by an arterial blood gas remains a significant limitation of all pulse oximetry. Clinician awareness of technological limitations of pulse oximetry, because of the device and the underlying pathophysiology of the patient, is crucial for appropriate use.^9,10,19–21  Portable pulse oximetry technology may connect the silos between respiratory care, subspecialists, and general care providers and help cross these barriers.

The new NHLBI guidelines regarding the diagnosis and management of asthma have a series of recommendations about use of fractional exhaled nitric oxide (NO).²²  In general, a high fraction of exhaled nitric oxide (FENO) is believed to be a clinically available biomarker representing nitrosative stress, a cause of long‐term lung damage. Nitrosative stress is classically caused by the reaction of NO with superoxide to form peroxynitrous acid, and by the reaction of NO with oxygen to form nitrite (through N₂O₃) and NO₂. is a toxic gas; nitrite (protonated to form nitrous acid) and peroxynitrite (protonated to form peroxynitrous acid) injure the airways in a pH‐dependent fashion.^23–26 

FENO is also generally high in conditions with increased eosinophilic (type 2) inflammation, like allergic asthma, but low in CF and in PCD.²⁷  Results can be used, in conjunction with other data, to assess children (school‐age and older) at the initial evaluation for wheezing and, to some extent, to monitor response to anti‐inflammatory therapy. Results are also beginning to be used to help decide which personalized interventions may work for children with severe asthma, both as a stand‐alone test and as a read‐out for an airway pH, reactivity, or bronchodilation challenge test.^23,24,26  It is not yet known whether interventions made based on FENO results in children will be valuable in preventing long‐term loss of lung function. Normative data for children above 4 years of age have been established.²⁸ 

It is important to recognize that FENO reflects airway biochemistry that is more complex than simply regulation of nitric oxide synthase expression and activity. It is affected by the airway microbiome, by airway pH, and by S‐nitrosothiol metabolism.^23,29,30  Therefore, changes in the microbiome, airway pH, and S‐nitrosothiol concentration can be reflected in changes in FENO. Acutely, these changes can be evaluated during inhaled buffer challenge tests to assess likely responsiveness to treatments.^26,31  Similarly, FENO can be affected by genetics; for example, in single-nucleotide polymorphisms (SNPs) for GSNO reductase.^{23,24,29,32,33} 

The knowledge gaps related to FENO involve a lack of complete mechanistic understanding and challenges in expanding its role. More studies are needed regarding the determinants of FENO; those commonly described in the literature often tell an incomplete or inaccurate story. More work is also needed to define the role of FENO and FENO challenge testing in clinical trials.

The cost and time of mechanistic studies to elucidate the determinants of FENO are barriers. Similar barriers exist in defining expanded roles of FENO testing beyond canonical use for evaluating eosinophilic (type 2) airway inflammation and likelihood of response to inhaled corticosteroids. Connecting the silos between subspecialists familiar with FENO and general care providers is also crucial. Normative data for children <5 years of age need further study and validation.²⁸  Finally, the long‐term implications of abnormal FENO with regard to lung function in adults is not yet known and is likely to vary with the cause of the abnormality. For example, antigen stasis in PCD leads to oxidative stress, which lowers FENO and injures the airway. In this case, low FENO probably will prove to be associated with long‐term decreased lung function.

The past 2 decades have been characterized by a growing interest in the use of "big data" approaches in the study of complex noncommunicable diseases. Many research groups have applied different analytical platforms to characterize profiles of genes, transcripts, proteins, metabolites, or microbiologic species associated with a specific clinical condition. Specific examples include exhaled breath condensate, challenge tests, and unbiased cluster analyses of multiple tests.^{6,26,31,34–37}  These studies enabled the characterization of pathogenetic processes underlying a respiratory disease or a subtype of disease. The findings of these studies pave the way to the development of therapies targeted at the specific underlying process in a truly personalized approach.

To accelerate the development of novel biomarkers, knowledge gaps must be addressed. Translation between the research and clinical settings remains a significant gap. The major challenge is represented by the transition from the characterization of overall discriminating profiles to the identification of one or more biomarkers. Ultimately, novel biomarkers will need full standardization of the analytical techniques applied, external validation of findings, and efforts toward the identification of specific and easy measurable biomarkers of disease to be used cost‐effectively in clinical practice.

FENO is a well‐known biomarker already validated for use in subspecialty clinical practice. The technique to measure FENO is easy and the results readily available. The measurement of FENO can contribute to the diagnosis of asthma. Expanded uses of FENO, beyond the diagnosis of asthma, are in development and may aid in the diagnosis and management of other pediatric lung diseases.

Genetic mechanisms underlie several neonatal and pediatric lung diseases, and early diagnosis and intervention may lead to improved outcomes. Genetic analyses can be used either to study specific conditions based on the suspected clinical diagnosis (Table 1) or screen for several diseases known to affect the lung using next generation sequencing diagnostic panels (Table 2). During the neonatal period, there is significant overlap in the clinical manifestations of heritable lung diseases, and the diagnosis is frequently difficult to confirm using clinical criteria alone. Multigene panels are now available to confirm the diagnosis of heritable lung diseases, to provide important information regarding the prognosis, and to guide therapy.³⁸  Broad implementation of diagnostic testing, genetic counseling, and timely referral to specialized centers are critical for early intervention to improve long term complications and lower disease burden.

Establishment of community‐specialty clinical partnerships will help enhance early diagnosis. Building multidisciplinary networks that involve diverse and complimentary professions including (but not limited to) physicians, nursing, respiratory therapy, genetic counseling, psychology, and scientists will promote advancement of all of these topics and further break silos. The continued development of international initiatives, such as the International Council for Respiratory Care and the International Health Committee of the American Thoracic Society, will help spread these advances and spread relevant education to rural and isolated areas throughout the world.

Genetic testing can be used to diagnose lung diseases in the neonatal period and during childhood to inform underlying pathophysiology, pathways activation, and ultimately to guide therapy. By treating the underlying pathobiological mechanisms, genetic‐based precision therapy results in better outcomes, slower decline in lung function, and a longer lifespan.³⁹ 

Unfortunately, treatment eligibility for specific genetic variants (eg, CFTR modulator therapy in CF) have been predominantly determined in non‐Hispanic white populations. Therapeutic interventions for genetic variants as well as the frequency of these variants and clinical significance have not been well studied in minority populations. This has led to significant disparity in healthcare delivery as patients of minority race and ethnicity are less likely to be eligible for genetic‐based therapy. Ancestry‐specific genetic biomarkers need to be determined in each ancestral group to ensure health equity and improve healthcare delivery in the community. However, many of these potential tools are expensive and labor‐intensive, which may continue to make them less accessible to people who live and receive care in resource‐poor areas.

Many pulse oximeters are less accurate in dark-skinned individuals, particularly when oxygen saturation levels are low. This can result in an overestimation of oxygen saturation, making patients seem healthier than they are and results in subpar treatment and outcomes. This can be remedied by improved research recruitment of broader populations during trials that evaluate devices and by improved calibration of sensors.⁴⁰  Evaluation of the usefulness of current and future biomarkers in underrepresented populations is necessary to reduce disparity.

Although much work remains to overcome the barriers between the development of novel biomarkers and their implementation into practice, advances continue to be made. Cutting‐edge techniques in genetic assessment and minimally invasive biomarker development benefit patients throughout the world. These techniques particularly benefit pediatric patients for whom early detection is crucial and who are less tolerant of invasive diagnostic procedures.

Although the topics discussed in this manuscript represent only a fraction of novel biomarkers relevant to pediatric clinicians, the gaps and barriers described affect the development of nearly all novel biomarkers (Table 3).

BPD

bronchopulmonary dysplasia

CF

cystic fibrosis

FENO

fraction of exhaled nitric oxide

NHLBI

National Heart, Lung, and Blood Institute

PCD

primary ciliary dyskinesia

SNP

single-nucleotide polymorphism