A mathematical model to predict protein wash out kinetics during whole-lung lavage in autoimmune pulmonary alveolar proteinosis

Cytokine profiles associated with disease severity and prognosis of autoimmune pulmonary alveolar proteinosis

BACKGROUND

Pulmonary alveolar proteinosis (PAP) is characterized by an abnormal accumulation of surfactants in the alveoli. Most cases are classified as autoimmune PAP (APAP) because they are associated with autoantibodies against granulocyte-macrophage colony-stimulating factor (GM-CSF). However, GM-CSF autoantibody levels are unlikely to correlate with the disease severity or prognosis of APAP.

METHODS

We collected clinical records and measured 38 serum cytokine concentrations for consecutive patients with APAP. After exclusion of 21 cytokines because of undetectable levels, 17 cytokine levels were compared between low and high disease severity scores (DSSs). We also compared whole lung lavage (WLL)-free survival with cut-off values defined by receiver operating characteristic (ROC) curves of cytokine levels and WLL administration at 11 months.

RESULTS

Nineteen patients with APAP were enrolled in the study. Five were classified as DSS 1 or 2, while the others were classified as DSS 4 or 5. Comparison between DSS 1-2 and 4-5 revealed that the concentrations of IP-10 and GRO increased in the latter groups (p < 0.05). Fifteen patients underwent WLL. Comparison between those who underwent WLL within 11 months and the others showed that IP-10 and TNF-α were tended to be elevated in the former group (p = 0.082 and 0.057, respectively). The cut-off values of IP-10, 308.8 pg/mL and TNF-α, 19.1 pg/mL, defined by the ROC curves, significantly separated WLL-free survivals with log-rank analyses (p = 0.005).

CONCLUSIONS

The concentrations of IP-10 and GRO may reflect the DSSs of APAP. A combination of IP-10 and TNF-α levels could be a biomarker to predict WLL-free survival.