BACKGROUND: Although psychological interventions can be effective for the treatment of major depressive disorder, some patients' symptoms persist or rapidly recur after therapy. This study aimed to synthesize research findings on predictors and moderators of treatment response for persisting forms of depression, such as chronic, recurrent, and treatment-resistant depression. METHODS: A systematic review of studies investigating predictors and moderators of response to outpatient psychological treatment for adults with persisting forms of depression was conducted by searching Web of Science, Scopus, and PsycInfo. A total of 23 eligible studies were included, assessed for risk of bias, and summarized using a narrative synthesis. RESULTS: Sixty-five predictor and moderator variables were examined across studies, categorized into sociodemographic, clinical, interpersonal/personality, psychological, and treatment variables. Most variables were only examined in single studies, which were rarely adequately powered for predictor and moderator analyses. Among variables studied more frequently (age, gender, baseline depression severity, childhood trauma), only baseline depression severity was found to be a replicated and consistent predictor of poorer treatment response. Risk of bias was low to medium for the majority of studies. LIMITATIONS: Meta-analysis could not be done due to methodological heterogeneity among studies. CONCLUSION: Our current understanding of significant predictors and moderators for persisting forms of depression is limited. A high level of baseline severity of depressive symptoms is so far the only variable consistently associated with poorer treatment response in this clinical population.