fuid of patients with “idiopathic” PAP. Interestingly they were not identi ed in secondary or congenital PAP [16, 17]. When healthy non-human primates were exposed to GM-CSF autoantibodies puri ed from patients with PAP, they developed lung disease and had abnormal GM-CSF signaling [19, 20]. It was subsequently demonstrated that the presence of GM-CSF autoantibodies was pathognomonic for “idiopathic” PAP which prompted a nomenclature change from idiopathic to the more accurate term autoimmune PAP which accounts for the majority of cases of PAP syndrome [21]. GM-CSF autoantibodies are also found at low levels in healthy individuals so there appears to be a critical threshold at which antibody concentration impairs GM-CSF signaling resulting in disease [21]. Furthermore, the absolute level of GM-CSF autoantibodies does not correlate with disease severity [7, 65].

Lymphocytosis

Secondary Infections

PAP patients have an increased risk of infection by both common or opportunistic pathogens, including Nocardia spp., Mycobacteria, Aspergillus, and others [2, 18]. Secondary infections can occur at the initial clinical presentation or in the rst year or so and are associated with a higher mortality rate [72]. An increased infection risk is seen in mice in which the GM-CSF gene has been knocked out (Csf2KO), and these mice have increased susceptibility to a range of bacterial, fungal, parasitic, and mycobacterial pathogens. This increased infection risk can be explained in humans and mice with GM-CSF signaling de ciency based on the observation that macrophages and neutrophils exhibit numerous innate immune defects [18, 27, 37, 73–75].

Pulmonary Fibrosis

Lung pathology in autoimmune PAP includes pulmonary lymphocytosis, although this has not been well studied. Interestingly, GM-CSF also has regulatory effects on lymphocytes [66–68]. The BAL cell differential count reveals increased pulmonary lymphocytes composed of increases in both CD4+ and CD8+ cells [69]. Furthermore, variable in l- tration of B and T cell lymphocytes is typically present where pathological specimens are available [9, 70, 71]. While increased pulmonary levels of monocyte chemoattractant protein-1 (MCP-1) may contribute to pulmonary lymphocyte accumulation, the precise mechanism(s) responsible have not been de ned [67].

Clinical Manifestations

Clinical Presentation

Patients with PAP syndrome usually present with subacute onset of progressive dyspnea with or without cough, fatigue, scant production of whitish frothy sputum, and occasionally weight loss. The presence of fever and/or hemoptysis suggests the presence of secondary pulmonary infection. More than 30% of patients are asymptomatic. Diagnosis is often delayed by months or years due to the non-speci c symptoms and physical exam ndings; many patients are initially misdiagnosed with pneumonia or asthma. Physical examndings occasionally include crackles, or cyanosis however; digital clubbing is not a feature of autoimmune PAP but is occasionally seen in other PAP-causing diseases. The clinical presentation can be surprisingly benign compared to the severe presentation expected from the extent of the radiographic abnormalities typically seen in PAP, a nding that should raise the suspicion of PAP; a heightened awareness is an important aspect of the timely diagnosis of PAP.

Pulmonary brosis occurs to varying degrees among PAP-­ causing diseases; it is a less-common manifestation of autoimmune PAP, variably present in secondary PAP, and the predominant manifestation in congenital PAP although this manifestation appears to present later in life [76]. The pattern of brosis varies considerably and can present as non-­ speci c interstitial pneumonia (NSIP), usual interstitialbrosis, and can progress to end-stage brosis with honeycombing and/or traction bronchiectasis [76]. Thus, evaluation for the presence of brosis should be part of the routine follow-up of all patients with PAP [7]. While antibrotic agents are widely used in idiopathic pulmonary brosis [77, 78], they have only recently been licensed for progressivebrosing interstitial lung disease and thus could be used in patients with PAP if brosis is identi ed [79]. Lung transplantation remains the only viable treatment option for patients with signi cant brotic lung disease.

Diagnosis

Pulmonary Function Testing

Routine spirometry and lung volumes may be normal in patients with mild PAP or may show varying degrees of restrictive lung impairment that varies in proportion to lung disease severity. In contrast, the DLCO is a sensitive measure of PAP disease severity and is reduced in proportion to disease severity [2, 23]. The 6-min walk test is often normal in PAP patients with mild or even moderate disease. Arterial blood gases show a varying degree of reduction in the arterial PaO2 and increase in the alveolar-arterial difference in oxygen concentration (A-aDO2) both of which correlate with disease severity and thus are useful indicators of the need for treatment [2].