Changes in IGF-I, urinary free cortisol and adipokines during dronabinol therapy in anorexia nervosa: Results from a randomised, controlled trial

https://doi.org/10.1016/j.ghir.2015.07.006Get rights and content

Highlights

  • Leptin increased and urinary cortisol secretion decreased during dronabinol therapy.

  • The levels of bioactive IGF-I were negatively influenced by the amount of energy expenditure.

  • The circulating IGF-system was unaffected by the dronabinol induced weight gain.

  • Circulating adiponectin also remained unaffected by dronabinol treatment.

Abstract

Objective

Anorexia nervosa (AN) is characterised by complex neuroendocrine disturbances due to severe underweight, physical hyperactivity and purging behaviour. Cannabinoid agonists are used to palliate cachexia of various causes, but their interactions with the hormonal systems that are involved in energy metabolism have not been previously described in humans. Therefore we found it of interest to assess interactions between the synthetic cannabinoid agonist dronabinol and insulin-like growth factor I (IGF-I), urinary free cortisol (UFC) and adipokines in patients with chronic AN.

Design

This was a prospective, double-blind randomised crossover study, conducted at a specialised care centre for eating disorders. The results are based on twenty-four adult women with chronic AN, who completed the study. The participants received dronabinol (oral capsules, 5 mg daily) and matching placebo over four weeks, separated by a four-week washout period. Bioactive IGF was determined by a cell-based bioassay, whereas total IGF-I, IGFBP-2 and -3 and the two adipokines leptin and adiponectines were measured by immunoassays. The UFC excretion was determined by mass spectrometry.

Results

As previously reported, dronabinol treatment caused a small, yet significant increase in BMI as compared to placebo (+ 0.23 kg/m2; P = 0.04). This modest weight gain predicted a corresponding increase in bioactive IGF-I, while the amount of daily energy expenditure due to physical activity had a comparable but opposite effect. Nevertheless, neither IGF-I, bioactive IGF nor the IGFBPs levels changed significantly during dronabinol intervention as compared to placebo. Adiponectin also remained unaffected by the weight gain, whereas plasma leptin showed a transient increase at three weeks (P < 0.05). UFC levels were decreased during dronabinol intervention.

Conclusion

Our results showed that low-dosage therapy with the synthetic cannabinoid agonist dronabinol affected neither the concentration nor the activity of the circulating IGF-system in women with severe and chronic AN. However, our results suggest that such treatment may alleviate the increased hypothalamic–pituitary–adrenal axis activity seen in these patients.

Introduction

The cannabinoid system is suggested to promote orexigenic and metabolic effects via both central and peripheral actions [1]. Accordingly, during the past decades, synthetic cannabinoid receptor 1 (CBR1) agonists have emerged as a novel treatment approach of severe underweight [2], [3], [4].

In contrast to other somatic conditions associated with chronic cachexia, anorexia nervosa (AN) is a purely psychiatric disease associated with multiple and profound neuroendocrine disturbances. The long, often chronic and relapsing periods of severe underweight affect nearly all endocrine axes and results in a severe catabolic condition, which in combination with an increased protein turnover has been referred to as auto-cannibalism. Dronabinol – a synthetic cannabinoid agonist – has been associated with a minor weight gain in patients with chronic AN [5]. Intriguingly, levels of physical activity were also modestly increased [6], while the parameters assessing eating disorder related behaviour and symptomatology remained unchanged during treatment. These apparently contradictory observations suggest that dronabinol exerted metabolic rather than psychoactive effects. However, the potential impact of CBR1 stimulation on the impaired hormonal systems involved in energy homeostasis has not been previously investigated in this group of patients.

The regulation of IGF-I is a pivotal process in AN, both in adaptation to chronic starvation and in regenerative processes during nutritional restoration. Chronic AN patients develop a nutritionally acquired resistance to GH, resulting in a decreased hepatic production of IGF-I and low circulating levels of total, free [7] and bioactive [8] IGF levels. The CBR1 has been found in hypothalamic regions secreting somatostatin as well as GH releasing hormone, and a biphasic response of GH to tetrahydrocannabinol exposure was described in previous reports [9]. Furthermore, animal studies report that both cannabinoid agonists [10] and marijuana smoke condensate [11] induce a significant increase in IGF-I gene expression. However, the evidence regarding interaction between the cannabinoid system and GH–IGF-I axis in humans is extremely sparse.

Hypercortisolemia is present in many adults and adolescents with AN [12]. The anatomic cohabitation of the CBR1 and the corticotropin-releasing hormone (CRH) expression in the hypothalamus [13] supports the existence of a close interaction between the endocannabinoid system and the activity of the hypothalamic–pituitary–adrenal (HPA). In animal models, CBR1 signalling leads to a negative modulation of stress-induced HPA activation [14]. However, the evidence of such a relationship in humans remains limited [15].

The amount of adipose tissue is usually extremely low in patients with AN and consequently, circulating levels of adipose tissue-derived factors, commonly referred to as adipokines, are profoundly altered. Among these, leptin is the primary signalling hormone through which the hypothalamus senses the nutritional state and modulates food intake and energy balance accordingly [16], and indeed hypoleptinemia has been proposed to serve as a diagnostic marker of AN [17]. Being a long-term integrative signal of energy storage, leptin has been reported to increase with weight gain in AN patients [18] and conversely, to decrease with weight loss in obese subjects [19]. Besides BMI and body fat, levels of leptin are highly correlated with IGF-I [20] and urinary free cortisol (UFC) [21], suggesting that the physiological regulation of leptin is complex. Data from cannabis smoking HIV-infected adult men suggest that the secretion of leptin might also be modulated though the endogenous cannabinoid receptors [22]. In addition to leptin, adiponectin is an insulin-sensitising cytokine produced by adipose tissue, but in contrast to leptin, plasma adiponectin is normally decreased in obese subjects and elevated in lean subjects. However, in AN, adiponectin levels have been reported to be elevated [23], [24], reduced [25] or unchanged [26] in comparison to levels in normal-weight controls.

In vitro data suggest that endocannabinoids may stimulate adipocyte proliferation [27], leading to fat accumulation, independently of the amount of food ingested [27]. Furthermore, adiponectin decreased during CBR1 receptor stimulation in vitro [28], while it increased during CBR1 antagonist stimulation in vitro as well as in vivo [29], [30]. According to our current information, the effects of CBR1 stimulation on the endogenous adipokine secretion have not previously been documented in patients with AN.

The aim of the present study was to assess the pathways affected by CBR1 stimulation in patients with AN. To this end, we studied the circulating IGF-system and serum levels of leptin and adiponectin as well as the UFC following 4 weeks of treatment with dronabinol vs. placebo in patients with severe, long-standing AN.

Section snippets

Trial design

Twenty-four adult women with chronic AN participated in this double-blinded, randomised, controlled, crossover study conducted at Odense University Hospital in Denmark [5].

The participants attended both psychiatric and somatic therapy as inpatients or outpatients at our centre and fulfilled the diagnostic criteria for AN according to the Diagnostic and Statistical Manual of Mental Disorders (fourth edition, text revision) for at least five years.

The baseline therapeutic regime was kept

Results

The participants were Caucasian women with a median age of 28 (23.5–42.0) years and median disease duration of 10.5 (7–18) years prior to inclusion. Baseline BMI was 15.7 ± 1.7 kg/m2 and the patients weighted 74.7 ± 8.1% of their ideal body weight. Both body weight and the estimated energy expenditure due to physical activity were modestly increased during dronabinol compared to placebo, as recently described [5], [6]. Although modest, the change in body weight led to a statistically significant BMI

Discussion

The orexigenic effect of cannabinoid agonist therapy has been widely documented in patients with cachexia associated with cancer, AIDS and Alzheimer's [2], [3], [4]. However, the interactions between cannabinoid agonists and the IGF system and other main metabolic systems involved in energy metabolism have not previously been investigated in humans. This study investigated the impact of the synthetic cannabinoid agonist dronabinol on several hormone systems involved in energy homeostasis in

Conflict of interest statement

This research was carried out independently by the Department of Endocrinology, Center for Eating Disorders, Odense University Hospital and was monitored by The Regional Unit for Good Clinical Practice, being approved by The Danish Medical Agency (journal number: 2612–3699), The Regional Committee for Medical Ethics (project id: S-20080016), and The Danish Data Protection Agency. The authors followed the guidelines by the European Medical Writers' Association. To our knowledge, all previously

Acknowledgements

We thank the patients who participated in the trial. Charlotte Olsen and Lone Hansen, are thanked for helping with sample collection. Laboratory technicians Merete Møller, Hanne Pedersen, Susanne Sørensen, Kirsten Nyborg Rasmussen, Lone Kvist, Elsebeth Hornemann og Lisa Buus are thanked for helping with the hormone measurements. Claire Gudex is thanked for language editing of the final manuscript.

References (39)

  • A. Andries et al.

    Effect of dronabinol therapy on physical activity in anorexia nervosa: a randomised, controlled trial

    Eat. Weight Disord.

    (2014)
  • R.K. Stoving et al.

    Jointly amplified basal and pulsatile growth hormone (GH) secretion and increased process irregularity in women with anorexia nervosa: indirect evidence for disruption of feedback regulation within the GH-insulin-like growth factor I axis

    J. Clin. Endocrinol. Metab.

    (1999)
  • R.K. Stoving et al.

    Bioactive insulin-like growth factor (IGF) I and IGF-binding protein-1 in anorexia nervosa

    J. Clin. Endocrinol. Metab.

    (2007)
  • V. Rettori et al.

    Hypothalamic action of delta-9-tetrahydrocannabinol to inhibit the release of prolactin and growth hormone in the rat

    Neuroendocrinology

    (1988)
  • K.K. Miller

    Endocrine dysregulation in anorexia nervosa update

    J. Clin. Endocrinol. Metab.

    (2011)
  • U. Pagotto et al.

    The emerging role of the endocannabinoid system in endocrine regulation and energy balance

    Endocr. Rev.

    (2006)
  • S. Patel et al.

    Endocannabinoid signaling negatively modulates stress-induced activation of the hypothalamic–pituitary–adrenal axis

    Endocrinology

    (2004)
  • R.S. Goodwin et al.

    CB1 — cannabinoid receptor antagonist effects on cortisol in cannabis-dependent men

    Am. J. Drug Alcohol Abuse

    (2012)
  • V. Di Marzo et al.

    Leptin-regulated endocannabinoids are involved in maintaining food intake

    Nature

    (2001)
  • Cited by (20)

    • New therapeutic approaches to target gut-brain axis dysfunction during anorexia nervosa

      2019, Clinical Nutrition Experimental
      Citation Excerpt :

      After reporting the central and peripheral alterations observed during AN, we review here the potential effects of physical activity on gut-brain axis during refeeding, and propose future combined strategies for AN treatment. Activation of the HPA axis through high plasmatic levels of cortisol and corticotrophin-releasing factor (CRF) have been observed in AN patients [9], and decreased after treatment [10]. Recently, Scharner et al. reported activation of CRF in distinct brain nuclei, in female ABA mice [11].

    • Activity-based anorexia activates CRF immunoreactive neurons in female rats

      2018, Neuroscience Letters
      Citation Excerpt :

      A previous study reported elevated circulating cortisol levels in ABA rats [9], a finding likely associated with an activation of CRF signaling in the PVN and DMH based on the current findings. These alterations are possibly relevant in humans as patients with AN also display elevated serum cortisol levels [27] which decrease following successful treatment of AN [28]. Therefore, CRF might play a pathogenetic role in the development or maintenance of AN.

    • Activity-based anorexia activates nesfatin-1 immunoreactive neurons in distinct brain nuclei of female rats

      2017, Brain Research
      Citation Excerpt :

      It is to note that cortisol levels were reported to be increased in activity-based anorexia rats (Burden et al., 1993) as well as in patients with AN (Casper et al., 1979), a modulation potentially involving an increased signaling of brain nesfatin-1. Cortisol levels decreased after therapy of AN giving rise to a pathogenetic role of steroids (Andries et al., 2015). Activation of the HPA axis might also underlie the observed stimulatory effects of nesfatin-1 injected icv (Merali et al., 2008) or intraperitoneally (ip) (Ge et al., 2015) on anxiety.

    View all citing articles on Scopus
    View full text