ObjectiveTo observe the efficacy of recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) for pulmonary alveolar proteinosis (PAP).Materials and methodsA total of 55 patients with PAP were screened at Shanghai Pulmonary Hospital between May 2014 and May 2018. Among these, 42 were diagnosed with idiopathic PAP, 24 were included in this study, 20 were treated for 6 months, and 17 were followed up for additional 6 months. All patients received a subcutaneous injection of 75μg/d GM-CSF qd for 1 month. The therapeutic dose was adjusted according to the changes in the lesions of chest CT. If the lesions were absorbed, subcutaneous injections of 75μg/d GM- CSF qd and 75μg/d GM-CSF qod were given for 2 and 3 months, otherwise, the dose was increased to 150μg/d GM-CSF qd and 150μg/d qod for 2 and 3 months, respectively. All cases were treated once a day in the first 3 months and once every other day in the last 3 months. The total course of treatment was 6 months. After withdrawal, the patients were followed up for another 6 months. The deadline of follow up was September 30, 2019.ResultsTwenty patients completed the treatment and efficacy evaluation. One patient was completely cured, 16 cases improved, three cases were noneffective. After 1-month evaluation, 12 patients received an increased dose (150μg) from the second month of treatment. Seventeen patients completed the 12-month follow-up, among which fourteen improved. CT showed the lesions were slightly increased in three cases. Economic burden was the following: RMB 7324–15,190 Yuan were required for the 6-month treatment course, which is significantly lower compared to other treatment methods.ConclusionSubcutaneous injection of rhGM-CSF at low dose (75μg-150μg /d) is effective treatment for patients with idiopathic PAP.Trial registrationNCT01983657. Registered 16 April 2013.

Background: Idiopathic pulmonary fibrosis (IPF) needs a precise prediction method for its prognosis. This study took advantage of artificial intelligence (AI) deep learning to develop a new mortality risk prediction model for IPF patients.Methods: We established an artificial intelligence honeycomb segmentation system that segmented the honeycomb tissue area automatically from 102 manually labeled (by radiologists) cases of IPF patients’ CT images. The percentage of honeycomb in the lung was calculated as the CT fibrosis score (CTS). The severity of the patients was evaluated by pulmonary function and physiological feature (PF) parameters (including FVC%pred, DLco%pred, SpO2%, age, and gender). Another 206 IPF cases were randomly divided into a training set (n = 165) and a verification set (n = 41) to calculate the fibrosis percentage in each case by the AI system mentioned previously. Then, using a competing risk (Fine–Gray) proportional hazards model, a risk score model was created according to the training set’s patient data and used the validation data set to validate this model.Result: The final risk prediction model (CTPF) was established, and it included the CT stages and the PF (pulmonary function and physiological features) grades. The CT stages were defined into three stages: stage I (CTS≤5), stage II (5 < CTS<25), and stage III (≥25). The PF grades were classified into mild (a, 0–3 points), moderate (b, 4–6 points), and severe (c, 7–10 points). The AUC index and Briers scores at 1, 2, and 3 years in the training set were as follows: 74.3 [63.2,85.4], 8.6 [2.4,14.8]; 78 [70.2,85.9], 16.0 [10.1,22.0]; and 72.8 [58.3,87.3], 18.2 [11.9,24.6]. The results of the validation sets were similar and suggested that high-risk patients had significantly higher mortality rates.Conclusion: This CTPF model with AI technology can predict mortality risk in IPF precisely.