Endocrine Abstracts (2015) 38 OC1.1 | DOI: 10.1530/endoabs.38.OC1.1

A novel pharmacological approach to target LH and testosterone hypersecretion in women with polycystic ovary syndrome: a randomised, double-blind, placebo-controlled multi-centre randomised clinical trial of the neurokinin B receptor antagonist AZD4901

Jyothis George1, Rahul Kakkar5, Jayne Marshall4, Martin Scott5, Richard Finkelman6, Tony Ho5, Stuart McIntosh6, Johannes Veldhuis3, Karolina Skorupskaite2, Richard Anderson2 & Lorraine Webber4


1University of Oxford, Oxford, UK; 2University of Edinburgh, Edinburgh, UK; 3Mayo Clinic, Rochester, USA; 4AstraZeneca, Cambridge, UK; 5AstraZeneca, Waltham, USA; 6Former AstraZeneca Employee, NA, UK.


LH hyper-secretion, driven by increased GnRH pulsatility, underpins excess testosterone secretion – a key clinical feature of PCOS.

The kisspeptin-neurokinin B (NKB)-GnRH pathway has emerged as the pivotal regulator of reproduction. We hypothesised that pharmacologic blockade of NKB may address the central pathophysiology of LH hyper-secretion and hyperandrogenism in PCOS. We undertook a randomised, double blind, placebo-controlled multi-centre Phase II trial (Clinicaltrials.gov NCT01872078) of the NKB receptor antagonist AZD4901. Sixty-seven women (mean±S.D.) age 28±5 years, BMI 31.5±6.0 kg/m2, were randomised to receive 20, 40 or 80 mg/day of AZD4901 or placebo orally for 28 days.

Inter-group comparisons used mixed effects models for repeated measures with baseline values as a covariate. The primary endpoint was change in LH area-under-curve (AUC) between baseline and day 7, quantified over 8 h using 10-min LH sampling.

In the AZD4901 80 mg/day group, mean (geometric) LH AUC decreased from 67.4 (±1 S.D. limits 42.2, 108.0) at baseline to 36.0 (15.9, 81.6) IU/L*h at day 7, a baseline-adjusted decrease of 52% (95% CI 30–67%) relative to placebo (P=0.0003). Correspondingly, LH pulse frequency in this group decreased from 5.79 to 3.73 pulses/8 h, an adjusted mean reduction of 3.55 (95% CI 2.00–5.10%) pulses/8 h relative to placebo (P<0.0001). Total testosterone decreased from 2.16 (1.63, 2.87) at baseline to 1.55 (1.06, 2.27) nmol/l at day 7, an adjusted decrease of 29% (95% CI 14–41%) relative to placebo (P=0.0006).

Excluding presumed ovulators (serum progesterone >6 ng/ml at any study visit), all these endpoints remained significantly reduced at day 28. No statistically significant LH or T changes were observed with lower doses.

In summary, AZD4901 reduced serum LH, LH pulse frequency and serum testosterone in this first RCT to manipulate the kisspeptin-neurokinin B (NKB)-dynorphin system in PCOS. No statistically significant LH or T changes were observed with lower doses (Table S3); no drug-emergent serious adverse events were reported. In summary, in this first study to manipulate the kisspeptin-neurokinin B (NKB)-dynorphin system in PCOS, the NKB receptor antagonist AZD4901 specifically reduced serum LH, LH pulse frequency and serum testosterone. Current therapy of PCOS is mainly symptomatic: these findings present NKB antagonism as a potential therapeutic approach to treat the central neuroendocrine pathophysiology of this common clinical condition.