To the Editor:

It has been reported that autoantibodies against granulocyte/macrophage colony-stimulating factor (GM-CSF autoantibodies) cause autoimmune pulmonary alveolar proteinosis (PAP), which accompanies increases in serum markers such as KL-6 and SP-D (1). In a previous issue, Cummings and coworkers (2) reported two cases of PAP that occurred in indium-tin oxide (ITO)-processing workers. Of these, one showed elevated serum GM-CSF autoantibody levels; in the other case, diagnosis was made post-mortem and serum was not available. Based on these findings, the authors suggested the possibility that inhaled ITO induces GM-CSF autoantibodies, and therefore PAP. This hypothesis is intriguing, because it is well known that rheumatoid arthritis, a typical autoimmune disease, is associated with occupational dust exposure (3). Because the production of ITO has increased markedly in recent years, due to its use in transparent conductive coatings for liquid-crystal and plasma display panels, it is necessary to examine this hypothesis in industrial medicine.

In Japan, pathologically proven cholesterol granuloma and pulmonary fibrosis with elevated serum KL-6 levels have been reported to occur in subjects exposed to ITO (4–7). Re-evaluation of lung specimens revealed that localized PAP-like findings characterized by accumulation of periodic acid-Schiff–positive materials in alveoli with cholesterol deposits were present in four of seven cases, but major pathological changes consisted of cholesterol crystal–laden macrophages and fibrosis in the interstitium. In contrast, amorphous materials diffusely occupied the alveoli in the cases reported by Cummings and colleagues. Chest high-resolution computed tomography (HRCT) features demonstrated that the Japanese cases had interlobular septal thickening in all seven cases, but not typical ground glass opacity. Lung biopsy was performed between1 and 2 years after the beginning of ITO exposure in the cases reported by Cummings and coworkers, whereas biopsy was performed 4 to 20 years after ITO exposure in the Japanese cases. We speculate that alveolar proteinosis represents a relatively acute phase response, particularly in cases of massive exposure, and it is replaced by cholesterol granuloma and fibrosis in the long term.

To clarify whether inhaled ITO induces GM-CSF autoantibodies, we measured serum GM-CSF antibodies in 17 current and former Japanese ITO workers at two factories, all of whom showed high levels of KL-6 (age, 41 ± 11 yr; mean ± SD). Of these 17 workers, nine had interstitial abnormalities on chest radiography and/or HRCT. Work duration, serum indium concentrations, serum KL-6, and SP-D levels of the subjects were 10.5 ± 1.2 years, 34 ± 6 ng/ml (normal range, 0.3 ± 2.6), 1,067 ± 185 U/ml (< 500), and 117 ± 20 ng/ml (< 110), respectively. Serum GM-CSF autoantibody levels were below the cut-off level in all subjects. This is consistent with our observation that serum GM-CSF autoantibodies were not induced in rats chronically exposed to ITO. Thus, our cases suggest that ITO inhalation does not induce GM-CSF autoantibodies.

References

Section:

ChooseTop of pageReferences <<CITING ARTICLES