Pulmonary alveolar proteinosis (PAP) is characterized by the accumulation of surfactant phospholipids and proteins within the lung alveoli. Important advances have been made over the past 8 years in our understanding of this disease, offering new directions for research and patient care. First, genetically altered mice that are homozygous for a disrupted granulocyte-macrophage colony-stimulating factor (GM-CSF) gene developed a lung lesion with histologic resemblance to PAP. The surfactant is thought to be catabolized or cleared mostly by alveolar macrophages, this process being dependent on GM-CSF. Second, a neutralizing autoantibody against GM-CSF was found in serum and bronchoalveolar lavage fluid of patients with idiopathic PAP but not in healthy controls, thereby raising the suspicion that human PAP may be an autoimmune disease. The relationship between the antibody and disease pathogenesis remains unclear but data suggest that the GM-CSF antibody may have a potential role as a diagnostic test. No specific therapy exists for PAP. Sequential whole lung lavage is the standard of care. Exogenous therapy with GM-CSF may improve the lung disease in some patients with PAP but this therapy is still experimental. Interventions directed at treating a relative GM-CSF deficiency by administration of GM-CSF or lowering the antibody level (i.e. by plasmapheresis or immunosuppression) may hold promise as future therapy for this rare disease.