Breath tests (BTs) have been investigated as diagnostic tools to phenotype drug disposition in cancer patients in the pursuit to individualize drug treatment. The choice of the right phenotype probe is crucial and depends on the metabolic pathway of the anticancer agent of interest. BTs using orally or intravenously administered selective non-radioactive (13)C-labeled probes to non-invasively evaluate dihydropyrimidine dehydrogenase, cytochrome P450 (CYP) 3A4, and CYP2D6 enzyme activity have been published. Clinically, a (13)C-dextromethorphan BT to predict endoxifen levels in breast cancer patients and a (13)C-uracil BT to predict fluoropyrimidine toxicity in colorectal cancer patients are most promising. However, the clinical benefit and cost effectiveness of these phenotype BTs need to be determined in order to make the transition from an experimental setting to clinical practice as companion diagnostic tests.