There is a consensus that the use of comprehensive geriatric assessment (CGA) is good clinical practice for older patients with solid tumors or hematological malignancies. To be complete, a CGA must include a geriatric assessment and an intervention plan. According to the SIOG consensus, a CGA should assess several domains: functional status, comorbidity, cognition, mental health status, fatigue, social status and support, nutrition, and the presence of geriatric syndromes. Progress has been made in the definition of the best way to detect problems, but the benefits are mostly based on prognosis stratification and on the adaptation of cancer treatment. The present review aims to evaluate the level of evidence regarding geriatric interventions proposed following the detection of a problem in cancer patients in each domain mentioned in the SIOG consensus. An online search of the PubMed database was performed using predefined search algorithms specific for each domain of the CGA. Eligible articles had to have well-defined interventions targeting specific domains of the CGA. We screened 1864 articles, but only a few trials on single-domain interventions were found, and often, these studies involved small groups of patients. This review highlights the scarcity of published studies on this topic. The specific impacts of CGA-based interventions have not yet been demonstrated. Multi-domain interventions seem promising, especially when they are based on global assessments. However, standardization seems difficult considering the lack of evidence for each domain. New studies are necessary in multiple care contexts, and innovative designs must be used to balance internal and external validity. An accurate description of the intervention and what "usual care" means will improve the external validity of such studies.