Inhaled GM-CSF Therapy of Autoimmune PAP

Inhaled GM-CSF Reduces the Need for Whole Lung Lavage and Improves Gas Exchange in Autoimmune PAP Patients

This is a prospective, randomized, open-label, long-term, phase 2 study of inhaled granulocyte/macrophage-colony stimulating factor following whole lung lavage therapy in patients with autoimmune pulmonary alveolar proteinosis.

Autoimmune pulmonary alveolar proteinosis (aPAP) is a rare disorder of progressive surfactant accumulation and resulting hypoxemic respiratory failure caused by disruption of granulocyte/macrophage-colony stimulating factor (GM-CSF) signaling, which alveolar macrophages require to remove pulmonary surfactant. The current therapy of aPAP, whole lung lavage (WLL), is a procedure requiring general anesthesia, endotracheal intubation to isolate each lung, and mechanical ventilation of the untreated lung while the treated lung repeatedly filled with saline and drained while percussing the chest to loosen and emulsify the surfactant and saline to physically remove the excess surfactant. Inhaled GM-CSF is a promising pharmacotherapeutic approach shown in case reports, small series, moderate open-label studies, and two randomized, double-blinded, placebo-controlled trials to be safe and improve the clinical, physiological, radiological, and biochemical disease manifestations in patients with mild-moderate aPAP. In contrast to the present study, prior studies were too short in duration to permit an evaluation on the requirement for WLL, which aPAP patients require a mean of every 15 months. The present study addressed the effects on WLL by studying patients with moderate-severe aPAP and by utilizing a long-term follow up period. The study design included a screening visit (month -3) to establish eligibility, an observation period (-3 to 0 months) to establish the presence of progressive/unremitting aPAP and establish disease severity, a pre-WLL visit (-1 month), a baseline visit (month 0) during which all patients received a scheduled, baseline, bilateral WLL, a 10-month, open-label treatment period, and a 20-month follow-up period. Study visits were scheduled at months -3, -1, 0, 1, 3, 6, 10, 18, and 30 months. Patients were randomized by the statistician to the GM-CSF Group (n=9) or the Control Group (n=9). Investigators were blinded to group assignment until after the participant's baseline visit. Patients randomized to the GM-CSF group (n=9) received inhaled GM-CSF (sargramostim (Leukine®), 250 mcg daily every other week for 12 weeks beginning 1 week after the baseline WLL - termed GM-CSF induction therapy period, followed by a 4-week washout period during which no GM-CSF was administered), and then received inhaled GM-CSF (sargramostim, 250 mcg/day on days 1 and 3 of every 14-day period for 6 months - termed GM-CSF maintenance therapy period). Inhaled GM-CSF (Leukine®) was administered using AKITA2 APIXNEB nebulizer system (Activaero, Vectura GmbH, Germany). Patients randomized to the Control Group (n=9) received no further scheduled treatment. Any patient in either group experiencing with disease progression resulting in respiratory failure (defined as peripheral artery oxygen concentration (PaO2) <60 mmHg at rest or PaO2 >60 mmHg at rest AND a peripheral blood oxygen saturation (SpO2) < 90% OR a decline in SpO2 of 5% or more during exercise), received (unscheduled) rescue WLL and were considered to have failed their assigned intervention (GM-CSF or Control). The primary outcome measure was time, in months, between the scheduled baseline WLL and first administration of unscheduled 'rescue' WLL (termed 'time to rescue WLL'). Key secondary outcome measures included the response in peripheral arterial oxygen concentration (PaO2), alveolar-arterial difference in oxygen concentration (A-aDO2), diffusing capacity of the lungs for carbon monoxide (DLco), vital capacity, ground glass opacification (GGO) of the lungs measured by visual scoring of chest computed tomography (CT) scans, and serum biomarkers of PAP (carcinoembryonic antigen, Krebs von-Lungren antigen, Cyfra-21.1). Other outcome measures included the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36), serum GM-CSF autoantibody concentration, the peripheral white blood cell and platelet counts. The occurrence and timing of rescue WLL administration in each group was evaluated using Kaplan-Meyer analysis. The primary end point was analyzed as the difference in median time to rescue WLL between the GM-CSF group and the Control group. Categorical outcomes were compared using Fisher's exact test. Key secondary end points were evaluated using repeated measures analysis of variance (RM-ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary outcome measures were also evaluated by comparing the between-group mean (or median) values at each visit using Student's t-test (or Mann-Whitney test) after imputation of missing data using a last observation carried forward approach to reduce selection-type bias by comparing the corresponding group means or medians. All reported p values are two-sided and have not been adjusted for multiple testing. P values of less than 0.05 were considered to indicate statistical significance. Analysis of the primary and key secondary outcomes was performed with the use of Stata software version 14.2. Analyses of secondary outcome measures were performed with the use of Prism for Mac OS software, version 9.51. Anticipated results were intended to compare the effects of inhaled GM-CSF following baseline WLL to those of baseline WLL alone in patients with moderate to severe aPAP.

Scheduled Baseline WLL: All participants will receive scheduled bilateral WLL at baseline (month 0). GM-CSF induction treatment: All participants will receive inhaled GM-CSF (250 mcg daily, 7 consecutive days every other week for 12 weeks beginning 1 week after the scheduled baseline WLL). Washout period: All participants will not receive inhaled GM-CSF treatment for 4 weeks immediately following GM-CSF induction treatment. GM-CSF maintenance treatment: All participants will receive inhaled GM-CSF (250 mcg daily on days 1 and 3 of every consecutive 14-day period for 6 months beginning 17 weeks after the scheduled baseline WLL). Unscheduled Rescue WLL: Any participant experiencing progression of aPAP lung disease (defined as the disease progression resulting in respiratory failure (PaO2 at rest <60 mmHg of PaO2 > 60 mmHg at rest AND desaturation <90% at rest OR decline in SpO2 of 5% or more during exercise testing) will receive unscheduled rescue WLL.

Scheduled Baseline WLL: All participants will receive a scheduled bilateral WLL at baseline (month 0). Unscheduled Rescue WLL: Any participant experiencing progression of aPAP lung disease (defined as the disease progression resulting in respiratory failure (defined by a resting PaO2 <60 mmHg or > 60 mmHg and desaturation <90% at rest or decline in SpO2 of 5% or more during exercise testing) will receive unscheduled rescue WLL.

Time (in months) between the scheduled baseline WLL (at Study Month 0) and the first administration of an unscheduled rescue WLL (during the 30-months after Study Month 0)

Number of patients requiring an unscheduled rescue WLL during the 30-months immediately following administration of the scheduled baseline WLL therapy at Study Month 0.

Primary Analysis: Between-group difference in mean PaO2 at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in mean PaO2 at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, PaO2 is determined from lab values measured for an arterial blood gas procedure.

Primary Analysis: Between-group difference in mean A-aDO2 at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in mean A-aDO2 at each visit after imputation of missing data by the last-value carried forward method. For both of the primary and secondary analyses, A-aDO2 is calculated from lab values measured for an arterial blood gas, the ambient atmosphere pressure, and fraction of inspired oxygen at the time of the arterial blood gas procedure.

Primary Analysis: Between-group difference in mean DLCO at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in mean DLCO at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, DLCO is determined from pulmonary function tests.

Primary Analysis: Between-group difference in mean VC at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in mean VC at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, VC is determined from pulmonary function tests.

Primary Analysis: Between-group difference in median GGO Score at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in median GGO Score at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, the degree of severity of lung disease was calculated by determining the number of segments affected. The following scale was used: Grade 1 = 1 segment affected Grade 2 = 2 - 5 segments affected Grade 3 = 6 - 9 segments affected Grade 4 = 10 - 14 segments affected Grade 5 = > 14 affected segments

Primary Analysis: Between-group difference in median serum CEA levels at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in serum CEA levels at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, the serum CEA levels were determined via enzyme linked immunosorbent assay (ELISA).

Primary Analysis: Between-group difference in median serum KL-6 levels at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in serum KL-6 levels at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, the serum KL-6 levels were determined via enzyme linked immunosorbent assay (ELISA).

Primary Analysis: Between-group difference in mean serum Cyfra21.1 levels at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in serum Cyfra21.1 levels at each visit after imputation of missing data by the last observation carried forward method. For both of the primary and secondary analyses, the serum Cyfra21.1 levels were determined via enzyme linked immunosorbent assay (ELISA).

Analysis Method: Between-group difference in serum GMAb levels at each visit after imputation of missing data by the last observation carried forward method. The serum GMAb levels were determined via enzyme linked immunosorbent assay (ELISA).

Analysis Method: Between-group difference in WBC counts at each visit after imputation of missing data by the last observation carried forward method. The WBC counts were determined via complete blood count.

Analysis Method: Between-group difference in platelet counts at each visit after imputation of missing data by the last observation carried forward method. The platelet counts were determined via complete blood count.

Primary Analysis: Between-group difference in mean SF-36 General Health Score at each study visit after the scheduled baseline WLL was evaluated using repeated measures analysis of variance (ANOVA) after adjustment for baseline values, gender, age, and the number of patients at risk at each time point. Secondary analysis: Between-group difference in SF-36 General Health Score at each visit after imputation of missing data by the last observation carried forward method. The SF-36 General Health Score were determined from the RAND 36-Item Health Survey. Scoring the SF-36 questionnaire is a two-step process. First, pre-coded numeric values are recoded per the scoring key. All items are scored on a scale of 0 to 100 so that a high score defines a more favorable health state. Scores represent the percentage of total possible score achieved. In step 2, General Health is determined from the average score for questions 1, 33, 34, 35, and 36.

Inclusion Criteria: Female of male 18 years of age or older Diagnosis of autoimmune PAP Able and willing to provide written informed consent Eligible for whole lung lavage determined as the presence of persistent or progressive respiratory failure (PaO2 at rest < 60 mm Hg) or desaturation < 90% or > 5 percentage points during standard exercise Exclusion Criteria: Diagnosis with secondary PAP, congenital PAP, or hereditary PAP Contraindication to whole lung lavage Contraindication to administration of inhaled GM-CSF Chronic lung disease associated with already existing respiratory failure, such as emphysema or pulmonary fibrosis, chronic heart failure, ischemic heart disease, active pulmonary embolism, progressive cancer, and other severe metabolic conditions

Trapnell BC, Whitsett JA, Nakata K. Pulmonary alveolar proteinosis. N Engl J Med. 2003 Dec 25;349(26):2527-39. doi: 10.1056/NEJMra023226. No abstract available.

Beccaria M, Luisetti M, Rodi G, Corsico A, Zoia MC, Colato S, Pochetti P, Braschi A, Pozzi E, Cerveri I. Long-term durable benefit after whole lung lavage in pulmonary alveolar proteinosis. Eur Respir J. 2004 Apr;23(4):526-31. doi: 10.1183/09031936.04.00102704.

Wylam ME, Ten R, Prakash UB, Nadrous HF, Clawson ML, Anderson PM. Aerosol granulocyte-macrophage colony-stimulating factor for pulmonary alveolar proteinosis. Eur Respir J. 2006 Mar;27(3):585-93. doi: 10.1183/09031936.06.00058305.

Trapnell BC, Inoue Y, Bonella F, Morgan C, Jouneau S, Bendstrup E, Campo I, Papiris SA, Yamaguchi E, Cetinkaya E, Ilkovich MM, Kramer MR, Veltkamp M, Kreuter M, Baba T, Ganslandt C, Tarnow I, Waterer G, Jouhikainen T; IMPALA Trial Investigators. Inhaled Molgramostim Therapy in Autoimmune Pulmonary Alveolar Proteinosis. N Engl J Med. 2020 Oct 22;383(17):1635-1644. doi: 10.1056/NEJMoa1913590. Epub 2020 Sep 7.

Tazawa R, Ueda T, Abe M, Tatsumi K, Eda R, Kondoh S, Morimoto K, Tanaka T, Yamaguchi E, Takahashi A, Oda M, Ishii H, Izumi S, Sugiyama H, Nakagawa A, Tomii K, Suzuki M, Konno S, Ohkouchi S, Tode N, Handa T, Hirai T, Inoue Y, Arai T, Asakawa K, Sakagami T, Hashimoto A, Tanaka T, Takada T, Mikami A, Kitamura N, Nakata K. Inhaled GM-CSF for Pulmonary Alveolar Proteinosis. N Engl J Med. 2019 Sep 5;381(10):923-932. doi: 10.1056/NEJMoa1816216.