Verena Gotta (1), Marie-Luise Summerer (1), Sven Wellman (1), Andrea Zelmer (1), Christian Schindler (2) and Nicole Ritz (1)
(1) University of Basel Children’s Hospital, Basel, Switzerland, (2) Swiss Tropical and Public Health Institute in Basel, Switzerland
Objectives: To characterize the age-dependency and normal ranges of eight cytokines in healthy children (IL-1ra, IL-2, IL-4, IL-6, IL-10, IP-10, IFN-γ and TNF-α), as well as the age-dependency of cytokine release following antigen stimulation ex vivo.
Methods: Healthy children aged 0-12 years undergoing elective/diagnostic procedures were eligible for inclusion. Whole blood was stimulated in vitro for 24 hours using Staphylococcus enterotoxin B, Phytohaemagglutinin or Candida albicans or left unstimulated, and cytokine concentrations were measured using a Luminex bead-based multiplex assay (precision: 2-19%, lower limit of quantification LLOQ=3.2 pg/mL). Data was analyzed by censored mixed effect regression analysis in NONMEM version 7.3 to evaluate linear and non-linear relationships of cytokine levels with age. Between-subject variability (BSV) was assumed to be log-normally distributed, unless variability-misspecification was evident in visual predictive check diagnostic. Only details of unstimulated cytokine data is presented here.
Results: A total of 271 children were included (median age 5.2 years, IQR 3.4-7.8; 74% male). Generally, unstimulated samples were negatively correlated with age, while stimulated samples were mostly positively correlated with age. A high proportion of unstimulated samples was <LLOQ (IL-2, IL-4, IL-10, IL-6: ≥70%, IFN-γ: 39%, IL-1ra: 7%, TNF-α: 6%, IP-10: 3%). TNF-α showed the strongest age-dependency, with an exponential decline of 30%/year (‘half-life’ t1/2=2.3 years, baseline=19 pg/mL, lower asymptote=7 pg/mL, p<0.001). IL-1ra decreased by 7%/year (t1/2=9 years, p=0.008), while a lower asymptote could not be quantified. BSV was large (74% to 240%), and was only to a small part (3-7%) explained by age. An age-dependency could not be quantified for the other cytokines, but was suggested by increasing proportions of data below the LLOQ in older age groups. Residual unexplained variability was 19%.
Conclusions: In healthy children unstimulated levels of TNF-α and IL-1ra decrease during the first 12 years of life. The other cytokines did not show a clear association with age, but the detection of such an association may have been impeded by the large proportion of data below LLOQ. The importance of age-dependent levels of these two important pro-inflammatory and anti-inflammatory cytokines need to be further investigated. It may be relevant for choosing the optimal dose of immune-modulating drugs in pediatric patients.
Reference: PAGE 25 (2016) Abstr 5969 [www.page-meeting.org/?abstract=5969]
Poster: Drug/Disease modeling - Paediatrics