What is pulmonary alveolar proteinosis (PAP)?
PAP is a syndrome - not a single disease, characterized by progressive accumulation of surfactant in pulmonary alveoli that causes restrictive lung impairment, hypoxemic respiratory insufficiency and, in severe cases, respiratory failure and death. PAP was initially described by Rosen in 1958 and is now recognized to occur in a group of heterogeneous diseases usefully classified as Primary PAP, Secondary PAP, and Disorders of Surfactant Production.
Pulmonary surfactant composition, function, and homeostasis
Pulmonary surfactant is comprised of about 90% lipids (mostly phospholipids and a small cholesterol and other neutral lipids) and 10% surfactant proteins (SP-A, SP-B, SP-C, SP-D).
Normally, surfactant comprises a thin layer on the alveolar surface that helps maintain alveolar stability by reducing surface tension at the alveolar wall-liquid-air interface: without surfactant, this surface tension would cause alveolar collapse (as occurs in premature babies with insufficient surfactant production).
Surfactant homeostasis is normally maintained by balanced production of surfactant in alveolar epithelial cells and clearance by these cells and alveolar macrophages, each of which clear about half of the excess/used surfactant. Alveolar macrophages require GM-CSF stimulation for normal surfactant clearance (and other host defense functions).
Disruption of surfactant homeostasis results in the development of PAP and can through various mechanisms including 1) reduction in the ability of alveolar macrophages to clear surfactant, 2) reduction in the numbers of alveolar macrophages (and hence reduction in the clearance capacity of the tissue-resident alveolar macrophage population), or abnormal surfactant production (and accumulation).
In Primary PAP (autoimmune PAP, hereditary PAP – see below), the ability of alveolar macrophages to clear surfactant from the alveolar/lung surface is reduced, which results in accumulation of surfactant in alveoli. In autoimmune PAP, very high levels of antibodies against GM-CSF are present in the blood and lungs and completely neutralize the activity of GM-CSF. In hereditary PAP, GM-CSF signaling is disrupted by genetic mutations encoding alpha or beta subunits of GM-CSF receptors.
In Secondary PAP (many diseases), surfactant clearance by alveolar macrophages is also reduced (because either the numbers or functions of these cells are reduced), which results in the accumulation of surfactant in alveoli.
In Disorders of Surfactant Production (multiple diseases), abnormal surfactant production results in surfactant that is dysfunctional and causes parenchymal lung disease, which varies with the gene affected as well as the type of exact mutation within the effected gene(s) – the result is a variable degree of surfactant and significant parenchymal architectural distortion of the alveolar structures.
What diseases are associated with development of PAP syndrome?
Autoimmune PAP due to high levels of GM-CSF autoantibodies
Hereditary PAP due to CSF2RA mutations
Hereditary PAP due to CSF2RB mutations
Immune deficiency and chronic inflammatory syndromes3
Toxic inhalation syndromes5
Pulmonary Surfactant Metabolic Dysfunction Disorders
SFTPB mutations resulting in SP-B deficiency
SFTPC mutations resulting in SP-C dysfunction
ABCA3 mutations resulting in abnormal surfactant production
NKX2.1 mutations resulting in abnormal surfactant production
1 Includes: acute lymphocytic leukemia, acute myeloid leukemia, aplastic anemia, chronic lymphocytic leukemia, chronic myeloid leukemia, myelodysplastic syndromes, multiple myeloma, lymphoma, Waldenstrom’s macroglobulinemia.
2 Includes: adenocarcinoma, glioblastoma, melanoma.
3 Includes: acquired immunodeficiency syndrome, amyloidosis, fanconi’s syndrome, agammaglobulinemia, juvenile dermatomyositis, renal tubular acidosis, severe combined immunodeficiency disease.
4 Includes: cytomegalovirus, mycobacterium tuberculosis, nocardia, pneumocystis jirovecii (formerly carinii).
5 Includes: dusts (inorganic) – aluminum, cement, silica, titanium, indium; organic dusts – agricultural, bakery flower, fertilizer, sawdust; fumes – chlorine, cleaning products, gasoline/petroleum, nitrogen dioxide, paint, synthetic plastic fumes, varnish.
What is the epidemiology of PAP?
PAP is very rare - probably affecting less than 10,000 people in the United States, and occurs in men, women and children of all ages, ethnic backgrounds, and geographic locations.
Autoimmune PAP – caused by GM-CSF autoantibodies – is the most common type and accounts for about 85-90% of cases. It most commonly occurs in the 20 – 40 year age range but also in children as young as 3 years and the elderly as old as 90 years. Smoking is a known risk factor.
Hereditary PAP – caused by genetic mutations that disruption the function of GM-CSF receptor alpha or beta subunits – is very rare and may account for 5% of cases. It typically presents in children in the 2 – 6 year old age range but can present in adults as old as 35 years.
Secondary PAP – associated with a wide variety underlying clinical conditions or toxic inhalation syndromes – is also very rare and may account for approximately 5% of cases. For many of the associations, the relationship or causality to the clinical condition or exposure is uncertain. The clinical presentation of secondary PAP is determined by the occurrence of the underlying clinical condition or exposure.
Disorders of surfactant production – caused by mutations in genes critical to surfactant production (SFTPB, SFTPC, ABCA3, NKX2.1) – is very rare and may account for approximately 5% of cases. Recessive mutations in these genes result in the production of biochemically and functionally abnormal surfactant, which has widely differing secondary physiological and tertiary anatomical effects on lung structure and function and clinical manifestations – all depending on which gene is affected and the nature of the mutation(s) involved. Various degrees of abnormal alveolar surfactant accumulation can occur (i.e., PAP syndrome) in association with other, typically more significant anatomic, physiologic, and/or clinical abnormalities.
What are the clinical features of PAP?
Primary PAP (autoimmune PAP, hereditary PAP) presents in most patients as exertional dyspnea of insidious onset that progresses over time. A non-productive cough is common. Sputum production occurs in about 5% of patients and when it occurs, sputum is typically whitish frothy material. Hemoptysis and fever are rare unless infection is also present. In severely affected individuals, cyanosis is also present. Digital clubbing is not a feature.
Secondary PAP presents in the clinical context of another underlying disease and thus, the presentation varies depending on the nature, timing of development of the underlying disease. For example, in PAP caused by the development of myelodysplastic syndromes, the presentation can be similar to that of Primary PAP but occurs in the context of the hematologic disease. In contrast, in acute inhalation exposure to significant amounts of respirable silica (e.g., as may occur in sandblasters), acute pulmonary toxicity and can be accompanied by respiratory symptoms such as cough.
Disorders of Surfactant Production present differently and at various ages depending on which gene and specific mutation is affected. Recessive mutations in SFTPB, ABCA3, NKX2.1 can present with respiratory failure at birth. Mutations in ABCA3 can also present as progressive respiratory insufficiency in young children and adolescents. In contrast, SFTPC mutations can present as interstitial lung disease at various ages.
Primary PAP usually goes unnoticed until disease progression is advanced – i.e., not until after the amount of surfactant accumulation in alveoli is sufficient to displace enough inhaled air to reduce oxygen uptake to cause exertional (or resting) hypoxemia and dyspnea. This process appears to occur slowly over months to years in many (or most) patients and goes completely unnoticed in the early stages. As surfactant accumulation continues over time, disease severity proceeds from sub-clinical (no symptoms) to mild exertional dyspnea (breathlessness) to more severe exertional dyspnea and, finally, dyspnea at rest.
PAP is frequently diagnosed incorrectly as either asthma in children before a chest x-ray is obtained, or as pneumonia in children or adults after a chest x-ray is obtained. Further, the accurate diagnosis of PAP is usually delayed until after bronchodilators and/or several courses of ‘appropriate’ antibiotics have failed to yield clinical improvement, prompting diagnostic reconsideration and evaluation.
Importantly, the disease severity judged based on presence of highly abnormal chest x-ray findings (see below) is frequently out of proportion – more severe – compared to the patient's symptoms, This discordance of radiological and clinical findings can be very useful diagnostically in the early recognition and accurate diagnosis of PAP and underscores the importance of clinical awareness and a high degree of clinical suspicion.
The natural history of PAP caused by loss of GM-CSF signaling (autoimmune PAP, hereditary PAP caused by CSF2RA/B mutations) includes the development of secondary infections with common as well as opportunistic organisms. In addition, some patients also can develop pulmonary fibrosis by a mechanism that is not defined.
The clinical course in primary PAP is quite variable and ranges from spontaneous improvement or remission of symptoms to progressive deterioration, respiratory failure, and death. Many individuals continue to have symptoms and require treatment by whole lung lavage (see below). The overall 5-year survival has been reported to be approximately 95% with treatment and about 85% in without therapy.
Autoimmune PAP is associated with an increased risk of microbial infection at pulmonary and extrapulmonary sites by a range of common pathogens and opportunistic organisms. Review of the medically literature suggests that, historically, approximately 18% of the mortality associated with PAP is caused by infections. While recent experience suggests that infection-related mortality is far lower than this, serious infections involving either common or opportunistic microorganisms can occur at presentation or any time during the clinical course and require therapy.
How is PAP diagnosed?
The timely and accurate diagnosis of PAP requires a high degree of clinical suspicion. Importantly, while routine clinical practice procedures and tests can establish the presence of PAP syndrome, they cannot identify the disease responsible. For this, an algorithm for differential diagnosis of PAP syndrome and specialized tests are needed. Some of these are currently available through as part of clinical research diagnostic testing programs like the one available at the Translational Pulmonary Science Center Laboratory associated with the Rare Lung Disease Consortium.
Approach to diagnosis
PAP should be suspected in patients with dyspnea of insidious onset and typical radiologic findings – diffuse ground glass opacification and superimposed septal thickening.
The history may be unremarkable except for dyspnea or may identify the presence of intercurrent infection suggested by the presence of fever, purulent sputum, and/or hemoptysis.
The physical examination may be otherwise unremarkable or inspiratory crackles may be present in lateral and dependent portions of the chest. Digital clubbing is not a feature of Primary PAP but may occur in some diseases associated with Secondary PAP or Disorders of Surfactant Production.
Routine laboratory testing usually not usually helpful but may reveal an increase in serum lactate dehydrogenase (LDH). Serum LDH is increased in proportion to disease severity in PAP.
Pulmonary function testing is normal in many patients but may reveal restrictive lung impairment in advanced disease. Importantly, the DLCO is decreased in proportion to disease severity in patients with PAP. Arterial blood gas measurement typically reveals a decrease in PaO2 and widened alveolar-arterial gradient that both decline in proportion to disease severity.
A standardized (American Thoracic Society) six-minute walk test can be helpful in assessing the severity of PAP. At rest, the peripheral capillary oxygen saturation (SpO2) may be reduced in patients with mild disease but usually falls during the course the tests in patients with clinically significant PAP and is reduced at rest in patients with advanced disease.
The chest x-ray in Primary and Secondary PAP of hematologic origin typically reveals bilateral patchy air space disease that appears similar to pulmonary edema but without the other radiographic signs of left heart failure. Various other patterns can occur including mixed alveolar, interstitial, or nodular infiltrates, and asymmetrical or focal abnormalities. Adenopathy, cardiomegaly, and effusions are not features of PAP and thus, not typically seen in uncomplicated PAP. The chest x-ray in Disorders of Surfactant Production varies depending on the specific disease present. In newborns with SP-B deficiency (SFTPB mutations) or some ABCA3 mutations, the radiographic signs of respiratory distress syndrome are present. In children and adults with other ABCA3 mutations or dysfunctional SP-C (SFTPC mutations), the chest x-ray reveals the presence of interstitial lung disease indicative of parenchymal abnormalities and fibrosis. Notwithstanding, the diagnostic value of the plain chest x-ray is limited by its lack of specificity.
Conventional chest computed tomography (CT) scans in Primary PAP and Secondary PAP of hematologic origin typically show bilateral, diffuse consolidation with poorly defined margins. High-resolution chest CT, which is superior, shows diffuse, patchy areas of ground-glass opacification with sharply-defined, straight and angulated margins representing the boundaries of secondary lobules or lung lobes. Also present in most cases is a ‘lattice’ of fine overlapping lines that form 3- to 10-mm polygonal shapes coinciding with the edges of the ‘geographic’ areas of ground glass opacification. Superimposition of these two patterns gives an appearance that has been described as “crazy paving,” which is characteristic but not diagnostic of PAP. Notably, crazy paving also occurs in hypersensitivity pneumonitis, Pneumocystis jiiroveci pneumonia, minimally invasive adenocarcinoma, lymphangitic carcinomatosis, cardiogenic pulmonary edema, acute lung injury, lipoid pneumonia. Notably, high-resolution chest CT is superior to both conventional CT and chest radiography in the assessment of the pattern and distribution of abnormalities and may demonstrate lesions even when the radiograph is normal. In Secondary PAP caused by toxic inhalation syndromes, the radiographic features of PAP may be accompanied by other radiographic signs related to the specific material inhaled. In Disorders of Surfactant Production, the radiographic features relate more closely to parenchymal lung abnormalities than those of surfactant accumulation and can include radiographic findings of respiratory distress syndrome or interstitial lung disease.
Bronchoscopy with evaluation of bronchoalveolar lavage cytology is helpful in establishing the presence of PAP syndrome as the cause of the clinical, physiologic, and radiographic abnormalities. Importantly, no findings from bronchoscopy, bronchoalveolar lavage, or cytopathology are capable of identifying the specific disease responsible for PAP.
Lung histopathology including examination of specimens obtained by transbronchial biopsy or surgical biopsy are capable of establishing the presence of PAP syndrome. However, lung biopsies are not capable of identifying the specific disease responsible in patients with PAP. Because lung biopsies are associated with increased morbidity and simple blood tests are now available that can identify the specific PAP-causing disease in more than 90% of patients, lung should be used only if the results of other tests are inconclusive and examination of lung parenchyma is needed.
Establishing a diagnosis of PAP syndrome
The typical diagnostic workup of PAP should include a standard medical history – including a detailed review of pulmonary exposures, hematologic problems, and serious infection history, physical examination, pulmonary function testing - including DLCO measurement, a standardized six-minute walk test to measure exercise-induced peripheral capillary oxygen desaturation and exercise-induced dyspnea/fatigue, and, if indicated, a high-resolution chest CT.
Identifying the specific PAP-causing disease
After a diagnosis of PAP syndrome is established, it is important that differential diagnosis be undertaken to identify the specific PAP-causing disease.
In a previously healthy individual with a new diagnosis of PAP syndrome (see above), an abnormal serum GM-CSF autoantibody (GMAb) test (see below) is usually sufficient to establish the diagnosis of autoimmune PAP. This is because 85 – 90% of all cases of PAP are caused by autoimmune PAP and the GMAb test is 100% sensitive and specific for this diagnosis.
When the GMAb test is negative or the measured serum GM-CSF autoantibody concentration is indeterminate (i.e., a value between clearly normal (<3 mcg/ml) and clearly abnormal (>9 mcg/ml)), a functional test to diagnose impaired GM-CSF signaling is useful – such tests include the STAT5 Phosphorylation Index test or the CD11b stimulation Index Test (see below). An abnormal STAT5-PI or CD11b-SI test with an intermediate GMAb Test result can confirm a diagnosis of autoimmune PAP.
When the GMAb test is normal (i.e., GM-CSF autoantibodies are not abnormally increased) and a STAT5-PI or CD11b-SI Test is abnormal (i.e., GM-CSF signaling is impaired), a diagnosis of hereditary PAP due to GM-CSF receptor mutations should be suspected. A variety of additional tests are available to further define the specific mutations responsible including: Serum GM-CSF, GM-CSF Ra, GM-CSF Rb, CSF2RA DNA/mRNA, CSF2RB DNA/mRNA, GM-CSF Clearance Tests (see below).
When the GMAb test is normal and the patient has been diagnosed with an underlying disease known to be associated with PAP, a diagnosis of Secondary PAP may be made. However, In such cases, genetic testing to exclude the presence of a Disorder of Surfactant Production may be needed.
In newborns, children, adolescents, and adults in whom a diagnosis of PAP syndrome has been established and, especially if parenchymal lung disease is also present, genetic testing for Disorders of Surfactant Production is indicated. These tests are available commercially.
For further information on these and other tests to identify surfactant related lung diseases, please contact the RLDC.
Specific blood tests used to identify PAP-causing diseases
Assessing disease severity
- Serum GM-CSF autoantibody (GMAb) Test – measures GM-CSF autoantibody concentration in serum by enzyme-linked immunosorbent assay (ELISA) using polyclonal GM-CSF autoantibody purified from individuals with autoimmune pulmonary alveolar proteinosis (aPAP) as the reference standard. Test results above the critical threshold value have a specificity of 100% and a sensitivity of 100% for a diagnosis of autoimmune PAP.
- Serum GM-CSF concentration (GM-CSF) Test – measures GM-CSF concentration in serum by an enzyme-linked immunosorbent assay (ELISA), using recombinant human GM-CSF as the standard. Serum GM-CSF is elevated in individuals with pulmonary alveolar proteinosis caused by defects in GM-CSF receptor function and in infection. The normal range for serum GM-CSF has not been reported. The sensitivity and specificity of this test for a diagnosis of hereditary PAP has not been reported.
- STAT5 Phosphorylation Index (STAT5-PI) Test – detects GM-CSF receptor signaling in blood leukocytes by measuring the level of GM-CSF stimulated phosphorylation of signal transducer and activation of transcription 5 (STAT5) using flow cytometry. Results are expressed as a STAT5 phosphorylation index (STAT5 PI) calculated as the mean fluorescence intensity (MFI) in stimulated cells minus that of un-stimulated cells divided by the MFI of un-stimulated cells multiplied by 100. The result is reported as ‘detected’ or ‘not detected’ for a normal or abnormal result, respectively. The specificity and sensitivity of this test for detection of GM-CSF signaling have not yet been reported.
- CD11b Stimulation Index (CD1b-SI) Test – detects GM-CSF receptor signaling in blood leukocytes by measuring the increase in cell surface CD11b on blood leukocytes stimulated by GM-CSF using flow cytometry. Results are expressed as a CD11b Stimulation index (CD11b SI) calculated as the mean fluorescence intensity (MFI) of stimulated cells minus that of un-stimulated cells divided by the MFI of un-stimulated cells multiplied by 100. The result is reported as ‘detected’ or ‘not detected’ for a normal or abnormal result, respectively. The specificity and sensitivity of this test for detection of GM-CSF signaling have not yet been reported.
- GM-CSF Receptor Alpha (GM-CSF R a ) Test - determines if GM-CSF receptor alpha is detected by antibody immunofluorescence staining on the surface of leukocytes in heparinized whole blood using flow cytometry. Normal and abnormal results are reported as ‘detected’ or ‘not detected’, respectively. The specificity and sensitivity of this test for detecting the GM-CSF receptor alpha chain has not been reported.
- GM-CSF Receptor beta (GM-CSF R b ) Test - determines if GM-CSF receptor beta is detected by antibody immunofluorescence staining on the surface of leukocytes in heparinized whole blood using flow cytometry. Normal and abnormal results are reported as ‘detected’ or ‘not detected’, respectively. The specificity and sensitivity of this test for detecting the GM-CSF receptor beta chain has not been reported.
- CSF2RA DNA Sequence Analysis (CSF2RA DNA) Test - determines a partial nucleotide sequence of the GM-CSF receptor α chain gene (CSF2RA) by standard methods. Sequence numbering is relative to the first base of the initiation codon (GenBank accession no. NM_006140.3 (alpha chain)).
- CSF2RB DNA Sequence Analysis (CSF2RB DNA) Test - determines a partial nucleotide sequence of the GM-CSF receptor β chain cDNA (CSF2RB) by standard methods. Sequence numbering is relative to the first base of the initiation codon (GenBank accession no. NM_000395 (beta chain)).
- CSF2RA mRNA Sequence Analysis (CSF2RA mRNA) Test - determines a partial nucleotide sequence of the GM-CSF receptor α chain cDNA (CSF2RA) by standard methods. Sequence numbering is relative to the first base of the initiation codon (GenBank accession no. NM_006140.3 (alpha chain)).
- CSF2RB mRNA Sequence Analysis (CSF2RB mRNA) Test - determines a partial nucleotide sequence of the GM-CSF receptor β chain cDNA (CSF2RB) by standard methods. Sequence numbering is relative to the first base of the initiation codon (GenBank accession no. NM_000395 (beta chain)).
- GM-CSF Clearance Test - measures the clearance of exogenously added recombinant GM-CSF by blood leukocytes from the patient maintained in culture media ex vivo. GMCSF is cleared rapidly by normal leukocytes in this assay but not by leukocytes from individuals with hereditary PAP caused by GM-CSF receptor dysfunction. The specificity and sensitivity of this test for a diagnosis of hereditary PAP have not been reported.
The presence or absence of symptoms (dyspnea), signs of infection (fever, hemoptysis), degree of impairment in oxygen uptake at rest and during exercise, are all used to determine the degree of lung impairment/ disease severity in patients with PAP. Specific assessments include pulse oximetry, pulmonary function testing to measure DLCO, six-minute walk testing, and arterial blood gas measurement. A scale of PAP disease severity based on blood gas measurement has been defined by Yoshikazu Inoue as follows:
• Stage 1 – PaO2 > 70 mm Hg, asymptomatic
• Stage 2 – PaO2 > 70 mm Hg, symptomatic (dyspnea, cough)
• Stage 3 – PaO2 < 70 mm Hg
• Stage 4 – PaO2 < 60 mm Hg
• Stage 5 – PaO2 < 50 mm Hg.
How is PAP treated?
Current ‘standard’ therapy
Whole Lung Lavage is currently considered to be the standard therapy for Primary PAP. It is also useful in patients with Secondary PAP of hematologic origin with compromised lung function who require urgent care for PAP. However, it provides little or no benefit in patients with Disorders of Surfactant Production. It is an invasive procedure performed under general anesthesia and separate endotracheal intubation of each lung, in which one lung is mechanically ventilated while the other is infused with large volumes (up to 50 L) of saline to physically “wash out” the accumulated surfactant lipids. While effective, the procedure has not been standardized across institutions with respect to the method (i.e., volume infused, use of mechanical percussion, the end point of an individual lavage procedure), indications for its use, methods for evaluating the treatment effectiveness, or timing of repeated procedures. Notwithstanding, it is widely held among practitioners to improve symptoms, radiographic findings, and gas exchange in PAP patients. Although WLL is safe in the vast majority of individuals, complications can include hypoxemia, pneumonia, sepsis, hydropneumothorax, and acute respiratory distress syndrome. The procedure is not performed in a patient with an active bacterial lung infection, since this can result in sepsis and shock. Bronchoscopic segmental or lobar lavage has been proposed as a safe alternative in patients in whom whole-lung lavage under general anesthesia is considered risky due to severe hypoxemia. Other alternatives include performance in a hyperbaric chamber, and use of complete cardiopulmonary bypass.
In practice, the indications for WLL therapy include dyspnea, exercise intolerance, and a desire to reduce the requirement for supplemental oxygen therapy. Reasonable indications for performing the procedure may include dyspnea limiting activities of daily living, arterial PO2 less than 60 mm Hg while breathing room air, significant desaturation (>5%) on exercise, and a shunt fraction greater than 10% to 12%.
Experimental GM-CSF augmentation therapy
A promising potential therapy currently in clinical research testing for patients with autoimmune is the aerosol administration of recombinant human GM-CSF (rhGM-CSF). Several early or preliminary studies show that inhaled GM-CSF may have a therapeutic effectiveness in between 62 and 95 percent of patients. However, formal toxicology studies have not been previously done to evaluate administration of rhGM-CSF by the aerosol inhalation route. This and other therapeutic approaches are being evaluated by clinical investigators of the Rare Lung Diseases Consortium.