Efficacy extrapolation and dose rationale for duloxetine in paediatric chronic pain
Silvia Grandoni (1,2), Paul Healy (1,2), Oscar Della Pasqua (1,2)
(1) Clinical Pharmacology and Therapeutics group, University College London (UCL), London, UK, (2) Consiglio Nazionale delle Ricerche (CNR), Rome, Italy
Objectives: It is broadly recognised that children of all age groups can perceive pain and there is an absolute need for safe and effective treatments. Presently available treatment options for acute pain in children include NSAIDs, acetaminophen and opioids [1]. Recently, tapentadol has been approved in the EU for acute pain treatment in children; its dual opioid-agonist and norepinephrine reuptake inhibitor (NRI) mechanism ameliorates the risk of opioid-typical side effects [2].
The situation is quite different for chronic pain, defined as persistent or recurring pain or pain lasting longer than 3 months [3]. Currently, there are no approved drugs for the treatment of chronic pain with only a few medications included in paediatric investigation plans for this condition [1].
Duloxetine is a serotonin and norepinephrine reuptake inhibitor (SNRI) approved by FDA for the treatment of fibromyalgia in adolescents (13-17 years). It is hypothesised that it may also show efficacy in paediatric patients with chronic pain, with the potential to reduce the risk of side effects associated with prolonged use opioids. Given the evidence of efficacy and good tolerability in adolescents, extension of its use in younger paediatric patients may be achieved by an extrapolation approach assuming similar PKPD relationships across age groups. The dose rationale can then be based on a duloxetine regimen that yields comparable exposure, irrespective of age or body weight.
Duloxetine is approved in adults for fibromyalgia, chronic musculoskeletal pain and diabetic peripheral neuropathy as well as in adolescents for fibromyalgia at 60 mg dose (oral capsule formulation) [4]. Here we propose a model-based approach for the selection and optimisation of the dose for paediatric patients aged between 6 months and 12 years old. In addition, a prospective PK bridging study is proposed to characterise the systemic exposure to duloxetine, including a sparse sampling scheme.
Methods: Nonlinear mixed effects modelling was used to scale the PK of duloxetine assuming comparable exposure-response relationships between adolescents and children. We used the observed duloxetine exposure associated with therapeutic response in adolescents as a target for this investigation. A published population PK model describing the PK of duloxetine in patients aged 7-18 years old was modified to allow allometric scaling of disposition parameters (clearance and volume of distribution) and to account for age-related changes in younger subjects, as basis for the prediction of the PK profiles in children from 6 years down to 6 months of age [5,6]. The model was then used to simulate and evaluate exposure in children based on weight bands, according to a mg/kg regimen. Doses were capped at 60 mg once daily. AUCss and Cmax were used as parameters of interest during dose selection and optimisation procedures. Virtual cohorts were used to simulate different doses and dosing regimens with body weight based on NHANES and CALIPER databases. A clinical study with 24 valuable patients, stratified into 3 age groups (6 months-<2 years, 2-< 6 years, and 7-<12 years) was subsequently simulated to assess the performance of different sampling schemes, including up to 5 samples per patient, following administration of a new liquid formulation.
Results: From these initial analyses it emerged that efficacious exposure (AUCss,24h 767 ng/mL*h median, 230-2624 ng/mL*h 5th-95th percentiles) can be achieved following administration of 2.2 mg/kg in children with weight below 10 kg, 2 mg/kg in the range [10-20) kg, 1.8 mg/kg in the range [20,30) kg, 1.5 mg/kg in the range [30,40) kg and 60 mg above. The use of an oral solution, formulated at 30mg/5mL, was selected as the preferrable strength to ensure accurate dosing in young children. A prospective PK study with 5 samples per subject allowed the characterisation of primary and secondary parameters (CL, AUC and Cmax).
Conclusions: Our work illustrates how quantitative clinical pharmacology principles can be used during the design phase of a study to deliver a robust dose rationale. The use of five weight bands is required to ensure target exposure is achieved in this group of patients. These results form the basis for further development of a paediatric formulation, which is currently not available.
References:
[1] Eerdekens M, Beuter C, Lefeber C, van den Anker J. The challenge of developing pain medications for children: Therapeutic needs and future perspectives. J Pain Res 2019;12:1649–64. https://doi.org/10.2147/JPR.S195788.
[2] Eissa A, Tarau E, Beuter C, Radic T, Watson E, Sohns M, et al. Tapentadol for the treatment of moderate-to-severe acute pain in children under the age of two years. J Pain Res 2021;14:229–48. https://doi.org/10.2147/JPR.S269530.
[3] Tutelman PR, Langley CL, Chambers CT, Parker JA, Finley GA, Chapman D, et al. Epidemiology of chronic pain in children and adolescents: A protocol for a systematic review update. BMJ Open 2021;11:1–4. https://doi.org/10.1136/bmjopen-2020-043675.
[4] CYMBALTA- duloxetine hydrochloride capsule, delayed release Eli Lilly and Company n.d. https://uspl.lilly.com/cymbalta/cymbalta.html#pi (accessed March 17, 2023).
[5] Lobo ED, Quinlan T, Prakash A. Pharmacokinetics of orally administered duloxetine in children and adolescents with major depressive disorder. Clin Pharmacokinet 2014;53:731–40. https://doi.org/10.1007/s40262-014-0149-y.
[6] Holford N, Heo YA, Anderson B. A pharmacokinetic standard for babies and adults. J Pharm Sci 2013;102:2941–52. https://doi.org/10.1002/jps.23574.