Dose determination of morphine to investigate the analgesic effects of morphine and pregabalin in healthy subjects using a combined NONMEM model
Monir Bertayli (1), Jeroen Elassaiss-Schaap (1)
(1) PD-value B.V
Introduction: Chronic pain (CP) is highly prevalent with 20.4%, and 7.4% of the adult US population suffering from chronic pain and high-impact chronic pain respectively [1]. CP causes major discomfort: the health-related quality of life of CP patients is as bad as palliative cancer patients [2]. Opioids are currently the most effective class of analgesics [3]. But with them comes a plethora of adverse effects such as drug abuse liability, cognitive impairment and sedation. There is a strong desire for improved treatments with less side effects. One possible avenue is exploring novel drug combinations. As part of the QSPainRelief consortium our goal is to identify novel analgesic drug combinations that are complementary or synergistic with each other and superior to current treatments. In a planned randomized, double-blind, placebo controlled four-way cross-over clinical trial of 24 healthy volunteers, pregabalin will be combined with intravenous morphine doses. The efficacy of the combined morphine and pregabalin treatment will be assessed by multiple tests, among which the ‘cold pressor test’ where subjects will place their hand for 2 minutes in a 35°C water bath directly followed by a 1.0°C bath. A PD endpoint is the time in seconds from immersion to hand withdrawal from the water bath, termed as pain tolerance threshold (PTT).
Objective:
-
To determine the optimal morphine dose for the planned clinical trial based on power calculations using simulations from a combined pregabalin and morphine PK/PD NONMEM model.
Methods: The cold pressor PTT effect for pregabalin has been modeled previously [4]. Pregabalin PK can be described by a one compartment model with lag time and linear elimination kinetics and was linked to a PD turnover compartment with a baseline latency of 16.9 seconds to describe the cold pressor PTT. Dahan et al. modeled the PK and the PD of morphine in healthy volunteers, where the cold pressor test functioned as a measure of the analgesic effect [5]. Morphine PK was described by a two-compartment model. The PD is described by multiplying the baseline latency by the concentration of the effect compartment divided by the concentration causing a doubling of the baseline latency. The two NONMEM models were combined into a single NONMEM model to simulate both single treatment arms and combinations of 300 mg pregabalin at 0h with different intravenous morphine doses at 2h and 5h. For the morphine hand latency withdrawal data, a log-logistic distribution was assumed, was applied in line with the morphine study[5]. The applied probability function was transformed into the quantile function (inverse cumulative distribution function) to generate observations following a log-logistic distribution. The applied inverse function is as follows; IPRED=IPRED*(R/(1-R))**(1/WK), where WK represents the steepness coefficient of the log-logistic distribution and R a random number generated following a uniform distribution. For the power calculations a multivariate regression model was setup using data from simulations of 1000 trials of each 24 subjects. Since pregabalin and morphine employ different pathways each effect was assumed to be additive on a 1:1 basis.
Results: Different contrasts where examined through plots where clinicians also recommended clinical trial design based on the PD profile over time and safety concerns. A morphine dose of 3 mg at 2h and 7 mg at 5h + 300 mg pregabalin was chosen and had a power of 88% with a mean differences of 5 seconds on the cold pressor PTT compared to a morphine dose of 3 mg at 2h and 7 mg at 5h without pregabalin.
Conclusions: We provided dose recommendation for a planned clinical trial in healthy volunteers by combining two available NONMEM models. Once results become available the accuracy of the predicted results can be assessed and if needed adjustments can be made.
References:
[1] Zelaya, Carla E., et al. "Chronic pain and high-impact chronic pain among US adults, 2019." (2020).
[2] Fredheim, O. M. S., et al. "Chronic non‐malignant pain patients report as poor health‐related quality of life as palliative cancer patients." Acta Anaesthesiologica Scandinavica 52.1 (2008): 143-148.
[3] Majedi, Hossein, et al. "Assessment of factors predicting inadequate pain management in chronic pain patients." Anesthesiology and pain medicine 9.6 (2019).
[4] van Esdonk, Michiel J., et al. "Population Pharmacokinetic/Pharmacodynamic Analysis of Nociceptive Pain Models Following an Oral Pregabalin Dose Administration to Healthy Subjects." CPT: pharmacometrics & systems pharmacology 7.9 (2018): 573-580.
[5] Dahan, Albert, et al. "Benefit and risk evaluation of biased μ-receptor agonist oliceridine versus morphine." Anesthesiology 133.3 (2020): 559-568.