C39-09 A Practical Tool for Measuring the Severity of Lung Disease in Pulmonary Alveolar Proteinosis

Rationale Despite decades of clinical trials including several global, randomized, blinded, placebo-controlled studies, no disease-specific pharmacotherapy of PAP is approved for use in the US or Europe. Most studies (and an existing disease severity score (DSS)) are all based on the use of arterial blood gas measurement, which is uncomfortable for patients, technically challenging, and not preferred by regulators as a primary end point for clinical trial intended to support regulatory approval. Therefore, we developed a tool to measure PAP lung disease severity (PAP-DSS) that incorporates patient-reported and clinical information and evaluated it in patients with autoimmune PAP. Methods The PAP-DSS score comprises a weighted composite of patient-reported data (San Diego-Shortness of Breath Questionnaire) and quality of life (Saint George’s Respiratory Questionnaire (SGRQ)-Total score), lung function (diffusion capacity of carbon monoxide percent predicted), and therapy (supplemental oxygen use). Scores range from 0-20, with higher values indicating greater severity. PAP-DSS was compared to short form-36 questionnaire (SF-36) quality-of-life tool, the visual analogue scale (VAS) for dyspnea and fatigue, spirometry, lung volumes, and SGRQ sub-component domain scores. Assessments included normality testing, correlations (Pearson or Spearman) with other outcomes, and internal consistency of the weighted contributors (Cronbach’s α). Results Among the 26 patients with autoimmune PAP evaluated in this FDA-supported study, the PAP-DSS was 5.6±1.0 (mean±SEM) points (range=0.4-17.3). PAP-DSS correlated very well with 1) SF-36 physical function subcomponent (Figure, panel a), 2) SF-36 subcomponent scores including energy, general health and social functioning; 3) VAS dyspnea score, 4) VAS fatigue score; 5) spirometry (forced vital capacity [FVC% predicted], forced expiratory volume in one second [FEV1% predicted]); 6) vital capacity [VC% predicted], and 7) SGRQ component scores including impact, activity, and symptom domains (Figure, panel b). The PAP-DSS score had excellent internal consistency as reflected by a Cronbach’s α value of 0.89 (acceptable values are >0.70). The average internal correlation between individual components of the PAP-DSS (mean r = 0.67) was within the optimal range (0.15-0.80); each component correlated strongly with the total PAP-DSS (>0.6) and removal of individual components only modestly reduced α (range 0.82-0.89) indicating each contributes to overall reliability of the score. Conclusion The PAP-DSS score is anticipated to be well-accepted by patients, clinicians, researchers, and regulators because it is comprehensive, correlates well with multiple accepted measures of PAP lung disease severity, and reflects how patients feel and function without relying on the use of arterial blood gas measurement.