Random unstimulated pediatric luteinizing hormone levels are not reliable in the assessment of pubertal suppression during histrelin implant therapy

Abstract

Background: Gonadotropin-releasing hormone agonist (GnRHa)-stimulated luteinizing hormone (LH) is the standard hormonal assessment for both diagnosis and therapeutic monitoring of children with central precocious puberty (CPP). Use of unstimulated (random) LH levels may be helpful in diagnosis and has gained popularity in monitoring GnRHa therapy despite lack of validation against stimulated values. The objective of this investigation was to assess the suitability of random LH for monitoring pubertal suppression during GnRHa treatment.

Methods: Data from a multi-year, multicenter, open-label trial of annual histrelin implants for CPP was used for our analysis. Children meeting clinical and hormonal criteria for CPP, either naïve to GnRHa therapy or previously treated with another GnRHa for at least 6 months who were being treated at academic pediatric centers were included in the study. Subjects received a single 50-mg subcutaneous histrelin implant annually until final explant at an age determined at the discretion of each investigator. Monitoring visits for physical examination and GnRHa-stimulation testing were performed at regular intervals. The main outcome measure was pubertal suppression during treatment defined by peak LH < 4 mIU/mL after GnRHa stimulation.

Results: During histrelin treatment, 36 children underwent a total of 308 monitoring GnRHa stimulation tests. Unstimulated and peak LH levels were positively correlated (r = 0.798), and both declined from the first to second year of treatment. Mean ± SD peak LH level during therapy was 0.62 ± 0.43 mIU/mL (range, 0.06-2.3), well below the normal prepubertal mean. Mean random LH was 0.35 ± 0.25 mIU/mL (range, 0.04-1.5), 10-fold higher than the normal prepubertal mean. The random LH levels were above the prepubertal upper threshold (<0.3 mIU/mL) in 48.4% of all tests and in 88.9% of subjects at some point during therapy.

Conclusions: In contrast with GnRHa-stimulated LH, unstimulated LH values frequently fail to demonstrate suppression to prepubertal values during GnRHa therapy for CPP, despite otherwise apparent pubertal suppression, and are thus unsuitable for therapeutic monitoring.

Trial registration: ClinicalTrial.gov NCT00779103.

Figures

Figure 1

Scattergram of unstimulated versus peak GnRHa-stimulated LH levels during the histrelin implant therapy. Data from all follow-up visits (n = 308) during the first year of histrelin implant therapy (open circles) and the second through fifth year of therapy (solid circles) are shown. The horizontal dashed line represents the a priori peak GnRHa-stimulated LH threshold level of 4 mIU/mL defining adequate suppression. The vertical dashed line indicates the published 0.3 mIU/mL pubertal threshold for unstimulated LH levels [15].

Figure 2

Mean (± SE) unstimulated LH levels (0 minutes) at each visit over 5 years of histrelin therapy. The horizontal dashed line indicates the 0.3 mIU/mL pubertal threshold for unstimulated LH levels [15]. Note that the time scale is not proportionate. SE is not shown beginning with the 36-month visit due to small sample size, and values in the fifth year are not shown due to small sample size (3 or less).