Clinical Trials of Cancer-related Pain: Translational research linking genetic variation to pain experience and response to oral opioids

Cancer-related pain is a major clinical problem. The World Health Organisation recommends morphine first-line for moderate to severe cancer-related pain, however several other strong opioids are available. There is wide inter-individual variation in morphine response, in terms of analgesic efficacy and side effects. Switching to an alternative opioid such as oxycodone has become common clinical practice to improve outcomes. There has been growing interest in potential genetic factors behind such differences in clinical response. This thesis aimed to compare the overall response rates of morphine and oxycodone, explore the effects of opioid switching in non-responders, and to redefine opioid response phenotypes to allow predictive modelling. Data from two opioid response studies is presented: the first a prospective observational study of morphine in cancer-related pain (n=298), the second a large open-label randomised controlled trial of oral morphine versus oral oxycodone in cancer-related pain (n=200). Single nucleotide polymorphisms from 15 candidate genes were tested using sequence specific primer polymerase chain reaction. Principal Component Analysis (PCA) was used to mathematically define opioid response phenotypes. Multivariate regression analysis of Component scores was used to build predictive models from genetic and clinical variables. There was no difference between the response rates of oral morphine and oral oxycodone when used first-line in cancer-related pain. Analgesia and adverse reaction profiles were similar. Switching from morphine to oxycodone and vice versa improved outcomes in the majority of non-responders. Opioid response appears to have three main domains identified by PCA in the two studies: analgesic response, upper gastrointestinal adverse reactions and central adverse reactions. Morphine and oxycodone although had similar response phenotypes had different clinical and genetic factors contributing to each, suggesting different mechanisms. Future work includes collaboration to enable larger studies to be conducted and to investigate the potential role of metabolites in oxycodone response