The centile of telomere length from early adulthood until senescence is mainly set between early gestation and adolescence.1 Hyperinsulinemic androgen excess (HIAE) is the most prevalent endocrinopathy of adolescent girls and is frequently driven by an absolute or relative excess of fat.2 Hyperinsulinemic androgen excess in adolescence is associated with a higher risk for a broad range of endocrine, metabolic, and cardiovascular complications later in life.3
Hyperinsulinemic androgen excess is traditionally treated with an oral contraceptive (OC; even when contraception is not an issue) that silences the ovarian androgen production, generates a pharmacological rise of circulating sex hormone–binding globulin, and leads to a condition of hyperestrogenic anovulation with regular pseudomenses.2,3
Early and prolonged insulin sensitization may become an alternative treatment because it cannot only attenuate the androgen excess (as OCs do) but also has anti-aging effects, such as lowering total, abdominal, visceral, and hepatic adiposity, attenuating low-grade inflammation, reducing intima-media thickness, and normalizing ovulatory function and menstrual regularity.2 The latter benefits may require the addition of an intrauterine device for contraceptive purposes.
Differential telomere lengthening might allow insulin-sensitizing interventions to exert more long-term benefits compared with OCs in adolescent girls with HIAE.
We measured leukocyte telomere length (LTL) by quantitative polymerase chain reaction in nonobese girls with HIAE who participated in a first randomized trial (OC vs insulin sensitization; ISRCTN45546616) across 24 months (18 months receiving treatment and 6 months not receiving treatment; Figure). This study was approved by the institutional review board of the Hospital Sant Joan de Deu, University of Barcelona, Spain. Participants provided written informed consent. The compared interventions were ethinyl estradiol cyproterone acetate vs a low-dose combination of 7.5 mg/d of pioglitazone, 62.5 mg/d of flutamide, and 850 mg/d of metformin (PioFluMet).4
Longitudinal results of leukocyte telomere length (by telomere repeat copy number to the single gene copy number [T/S] ratio) in adolescent girls with hyperinsulinemic androgen excess who were randomized to receive either an oral contraceptive (ethinyl estradiol cyproterone acetate [EE-CA]; N = 14) or an insulin-sensitizing treatment (N = 16) of low-dose pioglitazone (7.5 mg/d), flutamide (62.5 mg/d), and metformin (850 mg/d) (PioFluMet) between 0 and 18 months and who received no treatment between 18 and 24 months. The upper limit of the gray zone corresponds to a Z score of +2 in healthy control girls (mean [SE] age, 16.4 [0.4] years, mean [SE] body mass index [calculated as weight in kilograms divided by height in meters squared], 22.1 [0.2]; N = 18). Results in treatment groups indicate a slow on-treatment divergence followed by the rapid posttreatment convergence of telomere lengths.
aP = .01.
bP = .003 for differences between groups in changes from start (by 2-sided t test).
The LTL (expressed in the telomere repeat copy number to the single gene copy number ratio) remained unchanged during and after the OC intake. In contrast, the LTL doubled while taking PioFluMet and returned to baseline after the PioFluMet intake was stopped (Figure). The LTL changes across treatment groups during 18 months related inversely to fasting insulinemia, body fat fraction by dual energy x-ray absorptiometry, and visceral and hepatic adiposity by magnetic resonance imaging (all r values were between −0.53 and −0.57; all P values were between 0.002 and 0.007). The ratio of circulating neutrophils to lymphocytes was first similar in treatment groups and remained similar in and between groups across 24 months. Noteworthy adverse effects were not encountered in either treatment group.4
Prolonged insulin sensitization (with PioFluMet) is emerging as a first approach with antiaging effects and includes a slow marked reversible increment of LTL in adolescent girls with HIAE.
A comparably marked LTL increment was reported on initiating the treatment with sitagliptin in older Chinese adults with type 2 diabetes mellitus.5 In that study, telomere lengths were in the subnormal range at the start of treatment and increased to a healthy range within 2 months in parallel with improved glucose level control. In our young study population, telomere lengths were in the healthy range at the start of treatment and increased to the supranormal range after a longer intervention (12 to 18 months) in the absence of diabetes mellitus.
Future studies should disclose whether other insulin-sensitizing interventions (such as flutamide being replaced by spironolactone) can also elicit telomere lengthening in late adolescence and whether telomerase activity is up-regulated, either indirectly by a less adipose and more insulin-sensitive state or directly by components, such as pioglitazone, that can up-regulate the transcription of telomerase reverse transcriptase.6 Nevertheless, we may be closer to understanding why some insulin-sensitive women stay forever young.
Corresponding Author: Francis de Zegher, MD, PhD, Department of Development and Regeneration, University of Leuven, Herestraat 49, 3000 Leuven, Belgium (email@example.com).
Published Online: June 15, 2015. doi:10.1001/jamapediatrics.2015.0439.
Author Contributions: Dr de Zegher had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: de Zegher, Díaz.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: de Zegher.
Critical revision of the manuscript for important intellectual content: Díaz, Ibáñez.
Administrative, technical, or material support: Díaz.
Study supervision: All authors.
Conflict of Interest Disclosures: None reported.
de Zegher F, Díaz M, Ibáñez L. Association Between Long Telomere Length and Insulin Sensitization in Adolescent Girls With Hyperinsulinemic Androgen Excess. JAMA Pediatr. 2015;169(8):787-788. doi:10.1001/jamapediatrics.2015.0439