Background and purpose: Although the exact pathogenesis of idiopathic pulmonary fibrosis (IPF) is still unknown, the transdifferentiation of fibroblasts into myofibroblasts and the production of extracellular matrix components such as collagen, triggered by alveolar epithelial cell injury, are important mechanisms of IPF development. In the lungs of IPF patients, apoptosis is less likely to be induced in fibroblasts than in alveolar epithelial cells, and this process is involved in the pathogenesis of IPF. Experimental approach: We used a library containing approved drugs to screen for drugs that preferentially reduce cell viability in LL29 cells (lung fibroblasts from an IPF patient) compared with A549 cells (human alveolar epithelial cell line). Key results: After screening, we selected eperisone, a central muscle relaxant used in clinical practice. Eperisone showed little toxicity in A549 cells and preferentially reduced the percentage of viable LL29 cells, while pirfenidone and nintedanib did not have this effect. Eperisone also significantly inhibited transforming growth factor-β1-dependent transdifferentiation of LL29 cells into myofibroblasts. In an in vivo study using ICR mice, eperisone inhibited bleomycin (BLM)-induced pulmonary fibrosis, respiratory dysfunction, and fibroblast activation. In contrast, pirfenidone and nintedanib were less effective than eperisone in inhibiting BLM-induced pulmonary fibrosis under this experimental condition. Finally, we showed that eperisone did not induce adverse effects in the liver and gastrointestinal tract in the BLM-induced pulmonary fibrosis model. Conclusion and implications: Considering these results, we propose that eperisone may be safer and more therapeutically beneficial for IPF patients than current therapies.