Hypomethylating agents (HMAs) are widely used in patients with higher‐risk myelodysplastic syndromes (MDS) not eligible for stem cell transplantation; however, the response rate is <50%. Reliable predictors of response are still missing, and it is a major challenge to develop new treatment strategies. One current approach is the combination of azacytidine (AZA) with checkpoint inhibitors; however, the potential benefit of targeting the immunomodulator indoleamine‐2,3‐dioxygenase (IDO‐1) has not yet been evaluated. We observed moderate to strong IDO‐1 expression in 37% of patients with high‐risk MDS. IDO‐1 positivity was predictive of treatment failure and shorter overall survival. Moreover, IDO‐1 positivity correlated inversely with the number of infiltrating CD8+ T cells, and IDO‐1+ patients failed to show an increase in CD8+ T cells under AZA treatment. In vitro experiments confirmed tryptophan catabolism and depletion of CD8+ T cells in IDO‐1+ MDS, suggesting that IDO‐1 expression induces an immunosuppressive microenvironment in MDS, thereby leading to treatment failure under AZA treatment. In conclusion, IDO‐1 is expressed in more than one‐third of patients with higher‐risk MDS, and is predictive of treatment failure and shorter overall survival. Therefore, IDO‐1 is emerging as a promising predictor and therapeutic target, especially for combination therapies with HMAs or checkpoint inhibitors.