Mads Kreilgaard (1), Anne Brokjær (2), Ulrika S. H. Simonsson (3), Anne E. Olesen (2), Lona L. Christup (1), Albert Dahan (4), Asbjørn M. Drewes (2)
(1) Department of Drug Design and Pharmacology, University of Copenhagen, Copenhagen, Denmark, (2) Mech-Sense, Department of Gastroenterology & Hepatology, Aalborg University Hospital, Aalborg, Denmark, (3) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden, (4) Department of Anesthesiology, Leiden University Medical Center, Leiden, The Netherlands
Objectives: Morphine is the most commonly used opioid to alleviate moderate-severe pain [1]. For patients with gastrointestinal pain, rectal administration of morphine may be a feasible route to increase analgesic efficacy through central and local activation of opioid receptors. In order to select an optimal rectal dose for morphine in a forthcoming study, a population PKPD model was developed based on a pilot study in healthy volunteers undergoing experimental pain stimulation in order to select an optimal morphine rectal dose in a forthcoming study using simulations.
Methods: Ten healthy male volunteers were administered morphine, 2 mg IV or 10, 15, 20 mg rectally in a cross-over design on four different occasions. Ten blood samples were drawn in a sampling matrix design between 0-3 h for LC/MS/MS quantification of morphine in serum. During the same time period, the subjects were exposed to painful muscle pressure stimulation with a pneunomatic cuff algometer until moderate pain level (score 7 on a VAS where 5 is minimal pain and 10 unbearable pain). The PKPD model was developed in a sequential approach based on morphine plasma concentrations and muscle pressure tolerance scores over time, respectively, using NONMEM 7.2 and the First order conditional estimation method with interaction (FOCE-I). In the final model, PK and PD data were fitted simultaneously. Body weight was investigated as a covariate on the PK parameters. Model discrimination and evaluation were based on goodness-of-fit plots, visual predictive check, parameters precision, reduction in, objective function value and scientific plausibility. Based on the final PKPD model, simulations were performed in order to find an optimal dose matching criterias of ~ 15% peak response from baseline with morphine plasma levels
Results: A two-compartment model with first-order elimination best described the disposition of morphine. One Erlang-type absorption transit compartment [2] with first order rate constants (ka and ktr) best described the rectal absorption of morphine. The final PK model included allometric scaling of all clearance and volume parameters using body weight. The pain tolerance scores were best described by a direct log-linear model relative to morphine concentration. The model suggesting linear PK of morphine at rectal doses between 10-20 mg with a maximum increase in pain tolerance of 12% increased pain tolerance from baseline at the highest dose. Simulations from the PKPD model, suggested that ~15% median peak response could be obtained at a rectal dose of 30 mg, where the typical value for the predicted maximum morphine plasma concentration was 17 ng/mL (95% prediction interval: 6-36 ng/mL).
Conclusions: A non-linear mixed effect model was developed that adequately described morphine PK after IV and rectal administration, and the link to analgesia in healthy volunteers. The model predicted that a rectal morphine dose of 30 mg should provide target PK and PD response for a forthcoming study.
References:
[1] Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain (2006) 10(4):287-333.
[2] Rousseau A, Leger F, Le Meur Y, Saint-Marcoux F, Paintaud G, Buchler M, Marquet P: Population pharmacokinetic modeling of oral cyclosporin using nonmem: Comparison of absorption pharmacokinetic models and design of a bayesian estimator. Ther Drug Monit (2004) 26:23-30.
Reference: PAGE 23 (2014) Abstr 3093 [www.page-meeting.org/?abstract=3093]
Poster: Drug/Disease modeling - CNS