Gonadotropin hormone releasing hormone agonists alter prefrontal function during verbal encoding in young women

Summary

Gonadotropin hormone releasing hormone agonists (GnRHa) are commonly used in clinical practice to suppress gonadal hormone production in the management of various gynaecological conditions and as a treatment for advanced breast and prostate cancer. Animal and human behavioural studies suggest that GnRHa may also have significant effects on memory. However, despite the widespread use of GnRHa, the underlying brain networks and/or stages of memory processing that might be modulated by GnRHa remain poorly understood. We used event-related functional magnetic resonance imaging to examine the effect of GnRHa on verbal encoding and retrieval. Neuroimaging outcomes from 15 premenopausal healthy women were assessed at baseline and 8 weeks after Gonadotrophin Releasing Hormone analogue (GnRHa) treatment. Fifteen matched wait-listed volunteers served as the control group and were assessed at similar intervals during the late follicular phase of the menstrual cycle. GnRHa was associated with changes in brain response during memory encoding but not retrieval. Specifically, GnRHa administration led to a change in the typical pattern of prefrontal activation during successful encoding, with decreased activation in left prefrontal cortex, anterior cingulate, and medial frontal gyrus. Our study suggests that the memory difficulties reported by some women following GnRHa, and possibly at other times of acute ovarian hormone withdrawal (e.g. following surgical menopause and postpartum), may have a clear neurobiological basis; one that manifest during encoding of words and that is evident in decreased activation in prefrontal regions known to sub-serve deep processing of to-be-learned words.

Introduction

Over 3000 analogues of Gonadotropin hormone releasing hormone (GnRH) have been developed and tested (Karten and Rivier, 1986). These are primarily agonists (GnRHa) which down-regulate pituitary gonadotropin-releasing hormone receptors to produce a decline in ovarian hormone production. GnRHa is most commonly used in the management of benign uterine fibroids which are present in 20–30% of women over 30 years of age (Karten and Rivier, 1986). Pre-operative GnRHa administration has been shown to reduce intra-operative blood loss and facilitate more conservative operative techniques (e.g. reducing the need to carry out a midline incision) (Lethaby et al., 2001). GnRHa is also prescribed to facilitate super-ovulation during assisted reproduction (Pinkas et al., 2000) and to treat various other gynaecological conditions including endometriosis, (Shaw and Team, 1992) pelvic pain (Vercellini et al., 1993; Carey and Slack, 1996) dysfunctional uterine bleeding (Coulter et al., 1995), premenstrual syndrome (West and Hillier, 1994), and as a treatment for advanced cancer of the breast (Cheer et al., 2005) and prostate (Wirth and Froehner, 1999).

Recent research in animals (Casadesus et al., 2006) and humans (http://www.secinfo.com/d14D5a.z6483.htm, pp. 55–64) suggest that GnRHa may have beneficial effects on cognition, and Phase III trials are currently underway for the administration of GnRHa (as an adjunct to anticholinesterase drugs) in people with mild to moderate Alzheimer's dementia (http://www.centerwatch.com/trials/trial4063.html). However, most studies suggest that GnRHa has a negative effect on cognition (Varney et al., 1993; Newton et al., 1996). In particular GnRHa has been associated with reduced scores on neuropsychological tests of verbal episodic (Sherwin, 1996) and working (Grigorova et al., 2006) memory. Despite the widespread use of GnRHa (and the reported effects on cognition) the underlying brain networks and/or stages of memory processing that might be modulated by GnRHa remain poorly understood. The one in vivo brain imaging study into the effects of GnRHa on cognition (Berman et al., 1997) reported that GnRHa significantly reduced cerebral blood flow in the dorsolateral prefrontal cortex and the inferior parietal and temporal lobes during an executive function task in young women (most with premenstrual mood disorder). However, there have been no studies into the effects GnRHa on brain function during a memory task—despite this being the cognitive domain that has been most robustly reported to be affected by GnRHa (Sherwin and Tulandi, 1996). We therefore carried out a study to analyse the behavioural and functional effects of GnRHa on verbal episodic memory.

Prior studies of verbal memory have reported that successful encoding is associated with activation in the prefrontal cortex, particularly the left inferior frontal gyrus (Staresina and Davachi, 2006), the cingulate, medial temporal lobe and lateral temporal cortex (Fletcher et al., 2003). Studies have also reported that recognition success is associated with activation in the posterior parietal cortex (PPC), including inferior and superior parietal lobules, and medial structures extending from the precuneus to the posterior cingulate and retrosplenial cortices (Rugg and Yonelinas, 2003; Wagner et al., 2005). We hypothesized that GnRHa would impair verbal memory performance and that this would be associated with functional changes in these brain regions during successful encoding and recognition.