Extrapolating treatment to children less than 6 years old: the successful story of velmanase alfa in alpha-mannosidosis
Massimo Cella (1), Italo Poggesi (2), Marta Neve (2*), Maria Luisa Sardu (2)
(1) Chiesi Farmaceutici S.p.A., Parma, Italy, (2) Certara, Italy, *Affiliation when the analysis was conducted
Objectives: Alpha-mannosidosis (AM) is an ultra-rare disorder caused by the deficiency of the lysosomal enzyme alpha-mannosidase. It can manifest at a young age, with earlier onset associated with more severe disease progression. The effect of the enzyme deficiency is an accumulation of mannose-rich glycoproteins in the tissues, inducing impaired cell function, which in turn translates into different clinical manifestations, such as hearing loss, immunodeficiency, motor function disturbances, intellectual disabilities, skeletal abnormalities, and ataxia [1]. Velmanase alfa (VA) is a recombinant human alpha-mannosidase approved in Europe and USA as intravenous (IV) enzyme replacement therapy (ERT) of AM. Both approvals were based on a placebo-controlled study in patients 6–35 years of age. A small, open-label clinical study evaluated pharmacokinetics (PK), efficacy, and safety data in 5 patients 3-6 years of age.
The objectives of these analyses were:
- to develop a population PK model for VA,
- to describe the Exposure-Response (E-R) relationships between serum oligosaccharides (OGS) concentrations (clinical endpoint) and metrics of VA systemic exposure,
- to demonstrate that the dosing regimen of VA (1.0 mg/kg weekly) proposed for patients 6 years and older provides relevant clinical efficacy also in children < 6 years.
Methods: A population PK model was developed in NONMEM based on the VA PK data obtained from 7 clinical trials. Meaningful considered covariates were body weight, antidrug antibodies (ADA), dose, study, and age.
E-R exploration and modeling of OGS concentrations were performed with linear and non-linear models, using R (v4.0.2, nlme package v. 3.1-152). Predefined covariates for E-R analysis were ADA, sex, age, and body weight. PK and E-R models were extrapolated to children < 6 years to establish the clinical efficacy in this population via simulations using age, body weight, and sex for children (from 0 to 6 years of age) obtained from the Centers for Disease Control/National Health and Nutritional Survey database.
Results: The PK dataset included 468 measurable plasma PK observations from 39 subjects with AM (in most cases rolled over from one study to another, for a total of 78 PK occasions). At baseline there were 12 adults and 27 pediatric subjects. Response data included 289 OGS observations.
The PK model consisted of a 3-compartment open model with zero-order input and first order elimination from the central compartment. VA was characterized by low plasma clearance (CL) and volume of distribution at steady-state (0.401 L/h and 31 L, respectively). The PK model included the effects of body weight on CL terms and volumes (using standard allometric coefficients), the effect of study, ADA titer and dose on CL, and the effects of study and dose on the volume of the central compartment. Model parameters were estimated with good precision. Simulations indicated that the overall systemic exposure at the clinical dose of 1 mg/kg did not show clinically relevant dependencies on body weight and age.
E-R of OGS was described by an inhibitory Emax model. Typical values of baseline OGS, maximal effect (expressed as fractional inhibition from baseline) and VA concentrations providing 50% of the maximal effect (expressed in plasma) were 7.31 mM, 0.773, and 0.0694 μg/mL, respectively. None of the predefined covariates had significant influence on the E-R model.
AM in children 0-6 years of age is considered similar to the condition in older pediatric and adult subjects. In addition, similarity of VA pharmacology and response to intervention was indicated in the different age ranges based on the limited available data. This provided the rationale for the use of extrapolations to support the approval of VA in children <6 years of age. PK and E-R model-based simulations indicated that clinically relevant reduction of serum oligosaccharides was not dependent on body weight or age, thereby showing that VA at 1.0 mg/kg provides consistent, clinically meaningful therapeutic benefit in patients of all ages, including those from birth to 6 years of age.
Conclusions: These pharmacometric analyses were able to describe the PK and E-R of velmanase alfa for subjects with AM. Extrapolations based on PK and E-R models supported the use of velmanase alfa as ERT at the weekly dose of 1.0 mg/kg across all age ranges, including patients from birth to 6 years old.
References:
[1] Malm, D, Orphanet J Rare Dis, 2008 Jul 23;3:21