Implementing physics-based digital patient twins to tailor the switch of oral morphine to transdermal fentanyl patches based on patient physiology

Abstract

Fentanyl transdermal patches are widely implemented for cancer-induced pain treatment due to the high potency of fentanyl and gradual drug release. However, transdermal fentanyl up-titration for opioid-naïve patients is difficult, which is why opioid treatment is often started with oral/iv morphine. Based on the daily dose of morphine, the initial dose of the fentanyl patch is decided upon. After reaching a stable level of pain, the switch is made from oral/iv morphine to transdermal fentanyl. There are standard calculation tools for transferring from oral/iv morphine to transdermal fentanyl, which is the same for all patients. By considering the variations in the physiology of the patients, a unique switching strategy cannot meet the needs of different patients. This study explores the outcome in terms of pain relief and minute ventilation during opioid therapy. For this, we used physics-based simulations on a virtually-generated population of patients, and we applied the same therapy to all patients. We could show that patients' physiology, such as gender, age, and weight, greatly impact the outcome of the therapy; as such, the correlation coefficient between pain intensity and age is 0.89, and the correlation coefficient between patient's weight and maximum plasma concentration of morphine and fentanyl is -0.98 and -0.97. Additionally, a different combination of the duration of overlap between morphine and fentanyl therapy with different doses of fentanyl was considered for the virtual patients to find the best opioid-switching strategy for each patient. We explored the impact of combining physiological features to determine the best-suited strategy for virtual patients. Our findings suggest that tailoring morphine and fentanyl therapy only based on a limited number of features is insufficient, and increasing the number of impactful physiological features positively influences the outcome of the therapy.

Keywords: Opioid therapy; Pharmacodynamics model; Pharmacokinetics model; Tailored therapy; Virtual population.