Oxycodone is frequently used for treating cancer‐related pain, while not much is known about the factors that influence treatment outcomes in these patients. We aim to unravel these factors by developing a population‐pharmacokinetic model to assess the pharmacokinetics of oxycodone and its metabolites in cancer patients, and to associate this with pain scores, and adverse events. Hospitalized patients with cancer‐related pain, who were treated with oral oxycodone, could participate. Pharmacokinetic samples and patient‐reported pain scores and occurrence and severity of nine adverse events were taken every 12 h. In 28 patients, 302 pharmacokinetic samples were collected. A one‐compartment model for oxycodone and each metabolite best described oxycodone, nor‐oxycodone, and nor‐oxymorphone pharmacokinetics. Furthermore, oxycodone exposure was not associated with average and maximal pain scores, and oxycodone, nor-oxycodone, and nor‐oxymorphone exposure were not associated with adverse events (all p > 0.05). This is the first model to describe the pharmacokinetics of oxycodone including the metabolites nor-oxycodone and nor‐oxymorphone in hospitalized patients with cancer pain. Additional research, including more patients and a more timely collection of pharmacodynamic data, is needed to further elucidate oxycodone (metabolite) pharmacokinetic/pharmacodynamic relationships. This model is an important starting point for further studies to optimize oxycodone dosing regiments in patients with cancer‐related pain.