In the combined use of bronchodilators of different classes, ie, long-acting β2-agonists (LABAs) and long-acting muscarinic antagonists (LAMAs), bronchodilation is obtained both directly, through LABA-mediated stimulation of β2-adrenergic receptors, and indirectly, through LAMA-mediated inhibition of acetylcholine action at muscarinic receptors. The clinical trial data for LABAs/LAMAs in the treatment of chronic obstructive pulmonary disease (COPD) continue to be promising, and these combinations will provide the convenience of delivering the two major bronchodilator classes, recommended as first-line maintenance options in COPD treatment guidelines. COPD is a complex condition that has pulmonary and extrapulmonary manifestations. These clinical manifestations are highly variable, and several are associated with different responses to currently available therapies. The concept of a COPD phenotype is rapidly evolving from one focusing on the clinical characteristics to one linking the underlying biology to the phenotype of the disease. Identification of the peculiarities of the different COPD phenotypes will permit us to implement a more personalized treatment in which the patient's characteristics, together with his or her genotype, will be key to choosing the best treatment option. At present in Japan, fixed combinations of inhaled corticosteroids (ICSs) and LABAs are frequently prescribed in the earlier stages of COPD. However, ICSs increase the risk of pneumonia. Notably, 10%-30% of patients with COPD with or without a history of asthma have persistent circulating and airway eosinophilia associated with an increased risk of exacerbations and sensitivity to steroids. Thus, sputum or blood eosinophil counts might identify a subpopulation in which ICSs could have potentially deleterious effects as well as a subpopulation that benefits from ICSs. In this review, I propose one plausible approach to position ICSs and LABAs/LAMAs in clinical practice, based on both the extent of airflow obstruction and the presence of an asthma component or airway eosinophilic inflammation. This approach is a tentative move toward personalized treatment for COPD patients, and with progress in knowledge and developments in physiology, lung imaging, medical biology, and genetics, identification of COPD phenotypes that provide prognostic and therapeutic information that can affect clinically meaningful outcomes is an urgent medical need.