Purpose: To compare sociodemographic, health- and exercise-related characteristics of participants vs decliners, and completers vs drop-outs, in an exercise intervention during cancer treatment. Methods: Patients with newly diagnosed breast, prostate or colorectal cancer were invited to participate in a 6-month exercise intervention. Questionnaire, medical record and physical testing data were analyzed for between-group differences using independent t-tests and Chi2-tests. Results: Trial participants (n=577) were younger (59±12 yrs vs 64±11 yrs, p<.001), included more women (80% vs 75%, p=.012) and were scheduled for chemotherapy treatment (54% vs 34%, p<.001) compared to decliners (n=1391). A greater proportion had university education (60% vs 40%, p<.001), reported more anxiety (5.5±4.4 vs 4.3±3.7, p<.001), cancer-related fatigue, exercise self-efficacy (49.9±16.2 vs 45.0±20.1, p<.001), higher outcome expectations and less kinesiophobia (23.5±5.1 vs 24.5±5.0, p=.001) compared to decliners who completed baseline questionnaires (n=436). A greater proportion of participants were classified as ‘not physically active’, however, within the group who participated, more intervention completers were classified as ‘physically active’ at baseline compared to drop-outs. Completers also reported less kinesiophobia than drop-outs. Conclusion: Participants in a comprehensive exercise oncology trial differ in several sociodemographic, health and exercise-related aspects from those declining to participate, while differences between completers and drop-outs are less pronounced. A lack of health-enhancing exercise habits at baseline may be an incentive for accepting participation, while having such habits at baseline may increase the chances of completing exercise intervention trials. The findings provide useful insight for translating research results and planning future exercise interventions for cancer survivors.