Reduction in Alveolar Macrophage Size in Refractory Autoimmune Pulmonary Alveolar Proteinosis After Treatment With Pioglitazone

Autoimmune pulmonary alveolar proteinosis (aPAP) is a rare disorder characterized by alveolar accumulation of surfactant composed of lipids and proteins resulting in hypoxemic respiratory failure. Although spontaneous remission has been noted, whole lung lavage (WLL) is the standard of care in those requiring treatment. Pharmacologic treatments have been tried, most notably inhaled granulocyte-macrophage colony-stimulating factor (GM-CSF). Case series have shown improved gas exchange or pulmonary function parameters, and less need for WLL after therapy with inhaled GM-CSF.1–3 A more recent randomized trial however only showed minimal improvement in A-a gradient and no other significant benefits.4 Currently there are no Food and Drug Administraton approved pharmacologic treatments for aPAP. Peroxisome proliferator-activated receptor gamma (PPARγ) has been shown to play an important role in alveolar macrophage (AM) phospholipid and cholesterol efflux. PPARγ is present in the macrophages of healthy humans, and has been shown to be deficient in patients with aPAP and in an animal model of this disease which employs GM-CSF knockout (KO) mice.5–8 Upregulation of PPARγ leads to increased cholesterol efflux from AMs and decreased amounts of intracellular and extracellular phospholipids resulting in a decreased size of the AM compared with untreated mice.5 Administering PPARγ agonists to GM-CSF KO mice can reduce bronchoalveolar lavage (BAL) fluid turbidity and cholesterol levels,9 markers of disease severity in the mouse model. Although it is not known whether administering PPARγ agonists to humans with aPAP would lead to similar findings, PPARγ agonists, such as pioglitazone, may have potential in the clinical management of aPAP.5 We hypothesized that treatment of aPAP patients with pioglitazone would increase cholesterol efflux from AMs leading to a decrease in their size. We sought to measure the effect of pioglitazone on AMs in a patient treated with pioglitazone off-label for aPAP. CASE BACKGROUND A 52-year-old otherwise healthy male was diagnosed with aPAP in 2006 after presenting with hypoxic respiratory failure. A computed tomography scan showed a crazy paving pattern and BAL fluid demonstrated the typical milky appearance with Periodic-acid Schiff (PAS)-positive proteinaceous material on cytologic examination. Serum and BAL fluid were both positive for anti-GM-CSF antibodies (Mitogen Lab, Calgary, AB, Canada) confirming aPAP. He was treated with a WLL but was subsequently lost to follow-up until he returned with hypoxic respiratory failure again in 2009. He underwent several subsequent WLL in 2009 but continued to have refractory symptoms and hypoxemia and by January 2017 had undergone 42 unilateral WLL procedures using a technique similar to that described by Adbelmalak et al.10 Additional therapies included an 8-week course (24 sessions) of plasmapheresis as well as 8 months of inhaled GM-CSF. Neither produced any significant clinical improvement in pulmonary function testing or oxygen requirements. Despite the multiple WLLs by January 2017 the patient had gradually deteriorated and become oxygen dependent requiring 2 to 5 L via nasal prongs. There was diminishing benefit with subsequent WLLs as less proteinaceous material was removed compared with prior WLLs and improvement in oxygenation postprocedure was no longer noted. In addition, high resolution computed tomography scan of the lungs demonstrated progressive fibrosis (worsening reticulation with architectural distortion) with reduced ground glass and crazy paving. In view of limited options, progressive disease, and prior research showing the potential for reducing disease severity, a decision was made to treat with pioglitazone 30 mg daily. Institutional ethics were not obtained as it was a clinical decision to administer pioglitazone on an off-label, compassionate basis. Decision to pursue objective measurements pertaining to AMs was made after commencement of treatment to evaluate the effect of the medication. Patient consent was obtained for the off-label use of the medication as well as for the publication of this report. No adverse effects of the medication were noted while on treatment. METHODS BAL fluid was collected before initiation of pioglitazone therapy (January 2017) and after 9 months of treatment with no additional WLL performed in between. The fluid was centrifuged in a Shandon CytoSpin 3 cytocentrifuge (Thermo Fisher Scientific, Waltham, MA) at 1000 RPM for 3 minutes and cytospins prepared. The slides were stained with Wright-Giemsa stain and consisted mainly of AMs (∼99%). For assessment of mean AM diameter, a total of 75 AMs randomly selected in each cytospin were photographed ×200 magnification and then measured with ImageJ 1.51s software (https://imagej.nih.gov/ij/), blinded as to whether it was collected before or after treatment with pioglitazone. The longest diameter of the AM and the diameter orthogonal to this was measured. The average of these 2 measurements was considered the cell diameter. In addition, qualitative assessment of PAS-positive material and neutral lipids within the cell were performed. For these assessments, at least 100 AMs were assessed before and after pioglitazone treatment. Oil Red O was used to determine lipid content,5 and AMs were designated either positive or negative for stain. With respect to PAS, AMs were designated as having no/minimal, moderate, or strong positivity. A semiquantitative Histochemical score (H-score) was calculated for the PAS stain using methods previously described.11 A t test was performed on the mean diameters of the AMs before and after pioglitazone therapy, as well as the H-score before and after treatment (IBM SPSS Statistics 24.0.0.0). χ2 analysis was performed on the results of the Oil Red O staining. RESULTS The mean AM diameter decreased from 25.64 to 22.38 μm after therapy, representing a 12.7% decrease (P=0.002) (Fig. 1). There was a reduction in number of strongly PAS-positive AMs from 31.8% to 22.2% and increase in number of AMs with no/minimal PAS-positive material from 38.2% to 41.7%, but this did not meet statistical significance (Fig. 2). H-score for the PAS stain was 193.6 before pioglitazone, and decreased to 180.5 after treatment, however, this did not meet statistical significance (P=0.234) (Fig. 2). There was a nonsignificant reduction in proportion of AMs staining Oil Red O positive from 9.9% to 7.5% (P=0.427).FIGURE 1: Graphical representation of cell diameter prepioglitazone and postpioglitazone therapy. Each circle in the scatter plot signifies an individual cell size measurement. The horizontal line demonstrates the mean for the respective sample. AM indicates alveolar macrophage.FIGURE 2: Assessments of AM intracellular glycoprotein content. Clustered bar chart demonstrating fraction of AMs prepioglitazone and postpioglitazone therapy of different PAS stain grade. No statistical difference by H-score. AM indicates alveolar macrophage; H-score, Histochemical score; PAS, Periodic-acid Schiff.DISCUSSION AM size is a simple, although imperfect, marker of the total lipoproteinaceous material that has accumulated in patients with aPAP. We inferred that a significant decrease in the size of AMs may correlate with decrease in accumulation of lipid material within the cells and might therefore represent a decrease in disease severity. In this patient with refractory aPAP, a reduction in AM size was shown after 9 months of treatment with a PPARγ agonist. We believe the decreased AM size was due to improved cholesterol efflux from the cell as this would be consistent with previously published data of PPARγ agonist therapy in a mouse model of PAP.9 It should be noted that both prepioglitazone and postpioglitazone AM size was larger than the previously reported mean size for never-smokers (17.1 µm),12 consistent with aPAP. Despite the decrease in mean cell diameter, no significant change was seen in degree of intracellular PAS-positive material, or lipids seen with Oil Red O stain between samples. Previously rates of Oil Red O positivity of >90% in aPAP has been reported.5 The low proportion of cells staining positive for Oil Red O both pretreatment and posttreatment raises the possibility that the lipoproteinaceous material accumulation was not severe and therefore the fibrosis was the more clinically significant factor. There was neither clinical improvement nor deterioration in the patient’s oxygenation, pulmonary function testing, or subjective reporting of symptoms during this time frame. Starting therapy late after development of significant fibrosis may have minimized the potential for clinical improvement. Whether treatment earlier in his disease course would have led to clinical improvement is unknown. Similarly, it is unknown if pioglitazone prevented this patient from deteriorating further although this remains a possibility. Our paper has several limitations. Only one subject was evaluated and therefore we do not know if this could be reproduced in others. In addition, we do not know the expected changes in mean AM diameter on serial bronchoscopy and the decrease may have been due to normal intertest variability, although we feel this is less likely due to the number of AMs examined. The optimal dose of pioglitazone is not known for this situation, nor if a different dose would have yielded different results. Whether pioglitazone would lead to clinical improvement if used in other patients or at earlier stages of disease is not known. This medication’s use was off-label and no recommendations regarding routine use in aPAP can be made from this data. This report is the first to describe a reduction in AM size in a human patient with aPAP following the administration of a PPARγ agonist. Findings are consistent with prior laboratory and animal research and support the hypothesis that pioglitazone could potentially reduce disease severity in humans with aPAP. A phase 1 study of this drug in humans is underway (ClinicalTrials.gov Identifier: NCT03231033) and should give even more information about this drug as a potential therapy.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.