Objectives: Comprehensive Geriatric Assessment (CGA) has been incorporated into geriatric oncology to prevent unfavorable outcome from anticancer treatment. This study determined the value of CGA and medical oncologist's clinical judgment in predicting unfavorable outcome and explored whether treatment decisions can be based on CGA. Patients and Methods: In this prospective cohort study, a multidomain CGA was performed by a geriatric nurse and geriatrician in 110 consecutive patients aged ≥70 years, newly referred to a multidisciplinary oncology clinic. CGA domains included comorbidity, polypharmacy, mood, cognition, nutrition, functionality and physical performance. Medical oncologist's clinical judgment on expected tolerance of standard treatment was noted (N = 62). Unfavorable outcome was defined as any ≥grade three chemotherapy toxicity, dose reduction, postponement of treatment, death before start of treatment and early progression before first evaluation of treatment (N = 80). Results: CGA identified multidomain problems in 77 out of 110 patients (70.0%) and the medical oncologist had doubts about standard treatment tolerance in 30 out of 62 patients (48.4%). Unfavorable outcome occurred in 48 out of 80 patients (60%) who received anticancer treatment but could not be predicted by CGA, medical oncologists' clinical judgment or their combination. There was discrepancy between CGA and clinical judgment in 24 out of 62 patients (38.7%). Conclusion: Neither CGA, medical oncologist's clinical judgment or a combination could predict unfavorable outcome in our heterogeneous sample. CGA and clinical judgment did not align in more than one-third of patients.