(T-042) SimKid: An R package for Simulation of Virtual Pediatric Subjects

Andrew Santulli – Enhanced Pharmacodynamics, LLC; Sarah Cook – Enhanced Pharmacodynamics, LLC; Donald Mager – Enhanced Pharmacodynamics, LLC; Department of Pharmaceutical Sciences, School of Pharmacy and Pharmaceutical Sciences, University at Buffalo, The State University of New York, Buffalo, NY, USA; Scott Van Wart – Enhanced Pharmacodynamics, LLC

Objectives: Modeling and simulation for pharmacometrics often requires creation of virtual subjects, with distributions of model covariates and correlations between them that are reflective of the clinical trial or overall population characteristics. Body weight or body surface area (BSA) are commonly statistically significant and clinically relevant pharmacokinetic model covariates that are used for dosage calculations, especially during scale-down into pediatric populations. Creation of representative virtual subjects is paramount to accurate model-informed drug development when simulating models in pediatric populations. In this work, anthropometric growth chart data1-16 were incorporated into a pharmacometrics-oriented R package with the goal of facilitating the simulation of virtual pediatric populations.

Methods: Publicly available CDC, WHO, and Fenton growth chart data1-16 were collated into a standard format and saved as R data objects (.rda) for easy use and distribution. For CDC data for ages 2 to 20 years, weight-for-length LMS parameters were unavailable. Where the LMS parameters are the median (M), generalized coefficient of variation (S), and power in the Box-Cox transformation (L)17. Therefore, to ensure realistic correlations between weight and height, the correlation between z-score of height and z-score of weight was optimized by sex and 1-year age bin using a simulation approach with sum of squares criteria of fit between simulated and reported percentiles of BMI-for-age6. The LMS parameters for Fenton preterm growth of weight-for-gestational-age13,14 were also unavailable and were fit using a similar approach. The LMS parameters, respective equations used to obtain height and weight, optimized correlations between height and weight, and variability in z-scores were incorporated into an R package that allows for easy and rapid creation of virtual pediatric subjects. For each simulation, the user can specify the independent variables, such as the proportion of female subjects, the assumed distribution of age (uniform or truncated normal), and relevant age statistics (mean, standard deviation, and range).

Results: The SimKid R package can create virtual pediatric subjects with ages ranging from birth to 20 years. In order to validate that the virtual populations were reflective of anthropometric growth chart distributions and correlations, simulated percentiles of weight-for-age, length-for-age, and weight-for-length (for ages up to 2 years) or BMI-for-age (for ages greater than 2 years) were overlaid upon the respective observed percentiles obtained from anthropometric growth charts1-16. Validation figures confirmed that the SimKid package performed as expected, and a validation module was built into the package for user convenience and confidence. The package generates a data frame of virtual subject characteristics that includes age, sex, weight, height, BMI, and various calculations of BSA.

Conclusions: The SimKid R package simulates virtual pediatric subject demographics that are representative of real-world data based upon published growth chart data. Use of this R package can help simplify and potentially standardize the process of simulating virtual pediatric populations.

Citations: [1] National Center for Health Statistics. CDC Growth Charts: Weight-for-age for Children Birth to 36 Months. https://www.cdc.gov/growthcharts/data/zscore/wtageinf.csv. Accessed April 11, 2025.
[2] National Center for Health Statistics. CDC Growth Charts: Length-for-age for Children Birth to 36 Months. https://www.cdc.gov/growthcharts/data/zscore/lenageinf.csv. Accessed April 11, 2025.
[3] National Center for Health Statistics. CDC Growth Charts: Weight-for-stature for Children Birth to 36 Months. https://www.cdc.gov/growthcharts/data/zscore/wtstat.csv. Accessed April 11, 2025.
[4] National Center for Health Statistics. CDC Growth Charts: Weight-for-age for Children 2 to 20 Years. https://www.cdc.gov/growthcharts/data/zscore/wtage.csv. Accessed April 11, 2025.
[5] National Center for Health Statistics. CDC Growth Charts: Stature-for-age for Children 2 to 20 Years. https://www.cdc.gov/growthcharts/data/zscore/statage.csv. Accessed April 11, 2025.
[6] National Center for Health Statistics. CDC Growth Charts: BMI-for-age for Children 2 to 20 Years. https://www.cdc.gov/growthcharts/data/zscore/bmiagerev.csv. Accessed April 11, 2025.
[7] National Center for Health Statistics. WHO Growth Charts: Weight-for-age for Birth to 24 Months for Boys. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/growthcharts/WHO-Boys-Weight-for-age-Percentiles.csv. Accessed April 11, 2025.
[8] National Center for Health Statistics. WHO Growth Charts: Weight-for-age for Birth to 24 Months for Girls. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/growthcharts/WHO-Girls-Weight-for-age%20Percentiles.csv. Accessed April 11, 2025.
[9] National Center for Health Statistics. WHO Growth Charts: Length-for-age for Birth to 24 Months for Boys. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/growthcharts/WHO-Boys-Length-for-age-Percentiles.csv. Accessed April 11, 2025.
[10] National Center for Health Statistics. WHO Growth Charts: Length-for-age for Birth to 24 Months for Girls. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/growthcharts/WHO-Girls-Length-for-age-Percentiles.csv. Accessed April 11, 2025.
[11] National Center for Health Statistics. WHO Growth Charts: Weight-for-length for Birth to 24 Months for Boys. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/growthcharts/WHO-Boys-Weight-for-length-Percentiles.csv. Accessed April 11, 2025.
[12] National Center for Health Statistics. WHO Growth Charts: Weight-for-length for Birth to 24 Months for Girls. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/growthcharts/WHO-Girls-Weight-for-length-Percentiles.csv. Accessed April 11, 2025.
[13] Fenton 2013 2nd generation Preterm Growth Charts for Boys. https://ucalgary.ca/live-uc-ucalgary-site/sites/default/files/teams/418/fenton2013growthchartcolor-boys.pdf. Accessed April 11, 2025.
[14] Fenton 2013 2nd generation Preterm Growth Charts for Girls. https://ucalgary.ca/live-uc-ucalgary-site/sites/default/files/teams/418/fenton2013growthchartcolor-girls.pdf. Accessed April 11, 2025.
[15] Fenton TR & Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr. 2013;13:59.
[16] Fenton TR, Nasser R, Eliasziw M, Kim JH, Bilan D, Sauve R. Validating the weight gain of preterm infants between the reference growth curve of the fetus and the term infant. BMC Pediatr. 2013;13(1):92.
[17] National Center for Health Statistics. CDC Growth Charts: Definition of LMS parameters. https://www.cdc.gov/growthcharts/cdc-data-files.htm. Accessed April 23, 2025.

Keywords: R, Pediatrics, Pharmacometrics