Continuous reference intervals for pediatric testosterone, sex hormone binding globulin and free testosterone using...

Area of Science:

• Pediatric Endocrinology
• Biomarker Analysis
• Statistical Modeling

Background:

• Age-dependent reference intervals are crucial for interpreting testosterone (T), SHBG, free T (FT), and bioavailable T (BAT) in pediatric endocrinology.
• Traditional partitioning methods create abrupt age-group differences and data loss, especially during puberty.
• Existing continuous methods lack reproducibility due to complex parameter fitting.

Purpose of the Study:

• To present a reproducible workflow for non-parametric continuous reference intervals for T, FT, BAT, and SHBG.
• To address limitations of traditional age-partitioning methods in pediatric endocrine testing.
• To facilitate smoother, data-driven reference interval transitions across pediatric age groups.

Main Methods:

• Utilized the R language quantregGrowth package for non-parametric continuous reference interval analysis.
• Applied the methodology to testosterone (LC-MS/MS), calculated FT and BAT, and measured SHBG (Roche Cobas e601).
• Detailed the continuous interval methodology with code examples and illustrations for enhanced reproducibility.

Main Results:

• Developed and demonstrated a reproducible workflow for continuous reference intervals.
• Successfully applied the method to key pediatric endocrine hormones: T, FT, BAT, and SHBG.
• Provided detailed code and illustrations to ensure methodological transparency and ease of adoption.

Conclusions:

• The presented R workflow enables reproducible, non-parametric continuous reference intervals for pediatric endocrine tests.
• This approach overcomes the drawbacks of age-partitioning, offering smoother and more accurate interpretations.
• The detailed methodology supports wider adoption and consistent application in clinical and research settings.