ESPE Abstracts

Introduction: Salivary androgens represent a non-invasive marker of puberty that may have utility in population studies as well as in the clinical arena.

Objectives: To establish normal reference values of salivary androgens using LC-MS/MS and demonstrate the correlations between salivary androgens and pubertal development in boys.

Methods: School-based adolescent cohort study with two time points for collecting saliva samples two years apart. Five androgens (Testosterone;T, androstenedione; A4, 17-hydroxyprogesterone; 17-OHP, 11-ketotestosterone; 11-KT and 11β-hydroxyandrostenedione; 11OHA4 were analyzed in saliva samples using LC-MS/MS. In addition, self-reported assessment of puberty through the Pubertal Development Scale (PDS) was also collected at both time points. Receiver-operating characteristic (ROC) curves were used to determine the areas under the curves (AUCs), of each androgen as a predictor of self-reported voice maturation.

Results: A total of 1,166 saliva samples were available from 929 boys aged between 11-16 years at either baseline or follow up or both time points with the median age of 12.3 yrs (range 11.3-13.2) and 14.3 yrs (range 13.4-15.8) at baseline and follow up time point, respectively. Median salivary T increased from 7 pmol/l (10^th,90^th centile, 5, 41) in participants aged 11-12 yrs to 122 pmol/l (21.6, 267.4) in participants aged 15-16 yrs and median salivary A4 increased from 53 pmol/l (26.2, 92.0) in participants aged 11-12 yrs to 144 pmol/l (50.7, 241.2) in participants aged 15-16 yrs. In a subgroup analysis of 147 saliva samples that were collected within 90 days before or after PDS, salivary T and A4 concentrations showed the highest correspondence with self-reported voice-breaking (One-way ANOVA P < 0.005). ROC curve analysis showed that a salivary testosterone of 82.7 pmol/l and a salivary A4 of 113.4 pmol/l provided a sensitivity of 77% and 74%, respectively and a specificity of 76% and 74%, respectively. Salivary T concentrations revealed the highest linear correlation with salivary A4 (r = 0.75; P < 0.01) while the other androgens showed a lesser degree of correlation with salivary T with 17-OHP (r = 0.58; P < 0.01), 11-KT (r = 0.39; P < 0.01) and 11-OHA4 (r = 0.27; P < 0.01).

Conclusions: In boys aged 11-16yrs, an increase in salivary T and A4 is associated with self-reported pubertal progress including voice change. These two biochemical markers represent valid non-invasive biomarkers of puberty in population studies.