Abstract
Accumulating evidence describes the effects of oestrogen and other gonadal hormones on the central nervous system (CNS) and, in particular, on the mental state of women. An association between oestrogens and mental illness was initially recognized more than a century ago, whereby clinicians linked low oestrogen states (such as menstruation and menopause) to psychotic symptomatology [1]. More recent research from epidemiological and clinical studies has corroborated these early views and led to the oestrogen protection hypothesis, which proposes that oestrogen has a neuroprotective effect in women vulnerable to schizophrenia. Building on these clinical studies, animal studies and in vitro models have contributed to the understanding of the multiple physiological effects of oestrogens in the CNS.
According to the oestrogen protection hypothesis, high oestradiol production in young women, and changes in circulating oestradiol levels during a woman's life cycle, may contribute to the sex differences observed in the onset and course of schizophrenia. Epidemiological studies on the sex differences in schizophrenia suggest that women present with their first episode on average 4–5 years later than their male counterparts [2–5]. Additionally, life cycle studies have shown that women are more vulnerable for either a first episode or relapse of existing illness at two major periods of hormonal change: during the post-partum period (when oestrogen concentrations suddenly drop and return to normal levels) and during menopause (when a significant reduction in oestrogen levels is observed) [1], [6], [7]. Similarly, exacerbation of psychosis has been observed during low oestrogen phases of the menstrual cycle [8], [9].
Oestrogen's neuroprotective and psychoprotective actions may be mediated by a variety of routes ranging from rapid non-genomic actions, to slower genomic mechanisms, which may include permanent modification of neural circuits [10–14]. Oestrogen receptors are mainly located within the hypothalamic nuclei (important in oestrogen's role in sexual and reproductive behaviours), but have recently been identified in many brain areas including the amygdala–hippocampal area, substantia nigra and subthalamic nucleus, cerebellum and various areas of the cerebral cortex [12], [15], [16]. Via the oestrogen receptors, oestrogen modulates multiple neurotransmitter systems, including serotonin, dopamine, and norepinephrine [17–19]. It is therefore not surprising that widespread psychotherapeutic benefits of exogenous oestrogens have been described, in terms of improving neuronal health, elevating depressed mood, alleviating psychotic symptoms and enhancing cognition.
Several contemporary investigators have now reported promising findings using adjunctive hormones in the treatment of psychosis, mania and depressive symptoms, as well as the prevention of cognitive decline. Previous studies have indicated that patients receiving adjunctive transdermal oestradiol made a significantly greater improvement from acute psychotic symptoms, when compared to patients receiving adjunctive placebo [20], [21]. Due to the potential side-effects and risks of breast cancer, endometrial cancer and thromboembolic events associated with long-term, unopposed oestrogen treatment, selective oestrogen receptor modulators (SERMs) were developed as a hormone therapy for postmenopausal symptoms. SERMs, such as raloxifene, bind with high affinity to oestrogen receptors, and selectively exert either agonist or antagonist effects in a tissue-dependent manner [22]. Recent studies have suggested that raloxifene results in a reduced risk of cognitive impairment in healthy, postmenopausal women [23], [37]. It is of interest to note whether raloxifene may benefit cognitive functioning and/or psychopathology in postmenopausal women with schizophrenia.
The benefits of oestrogen may not be extended to women experiencing mania, rather anti-oestrogens may improve symptoms of mania. We found in the present sample of manic women (n = 4) that their symptoms worsened when given 50 µg transdermal oestradiol. Additionally, there have been case reports describing success in treating women with mania using an anti-oestrogen agent [24].
Tamoxifen, a SERM, is widely used as an anti-oestrogen agent in the treatment of breast cancer. In the CNS, tamoxifen acts as an oestrogen receptor antagonist, inhibiting oestrogen from carrying out its neuroprotective functions, as well as producing its own sequelae, which potentially includes depressogenic effects [25]. Additionally, tamoxifen is a potent inhibitor of protein kinase C (PKC) [26], which is believed to have anti-manic effects. PKC reductions are also found in mood stabilizers such as lithium and valproate [27]. Our pilot study indicated that over a 28 day trial period, women receiving adjunctive tamoxifen exhibited significant improvement in symptoms of mania, when compared to women receiving adjunctive placebo [28].
The following three case reports will demonstrate different applications of hormone modulation in the treatment of severe mental illness. These three women were all participants in large, double-blind, randomized controlled trials exploring hormone modulation conducted at the Alfred Psychiatry Research Centre. All participants provided written informed consent, according to the guidelines of the Australian National Health and Medical Research Council.
In each case, psychopathology symptoms were measured at baseline and then at weekly or fortnightly intervals using the Positive and Negative Syndrome Scale (PANSS) [29], which consists of three subscales measuring positive psychopathology (scores: 7 (nil symptoms) to 49), negative psychopathology (scores 7–49) and general psychopathology (scores 16–112) symptoms. Where appropriate, symptoms of mania were measured using the Clinician-Administered Rating Scale for Mania (CARS-M), scores range from 0 (nil symptoms) to 74 (severe mania) [30]; and symptoms of depression were measured using the Montgomery and Asberg Depression Rating Scale (MADRS), for which scores range from 0 (nil symptoms) to 60 (severe depression) [31]. Cognition was assessed at baseline and trial completion, using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) [32]. A Menstrual Cycle Questionnaire (MCQ) [33], [34] and hormone assays were used to stage the patient's menstrual cycle phase at baseline.
Case 1: adjunctive oestrogen treatment for psychosis.
The first subject was a 41-year-old caucasian woman with a diagnosis of schizoaffective disorder. This subject was recruited from the acute psychiatric inpatient unit to participate in an 8 week adjunctive oestrogen trial. When she commenced the trial her medications had been stable for approximately 2 weeks. These included: olanzapine (10 mg per day−1); paroxetine hydrochloride (40 mg per day−1) and diazepam (5 mg as required). The patient had stopped using the oral contraceptive pill 10 months earlier.
History
The subject initially presented in 1994, with significant psychotic symptoms including delusions and command hallucinations, as well as depression. She was subsequently diagnosed with schizoaffective disorder. In 1998, she disengaged the local psychiatric services and did not seek treatment for 8 years thereafter, but remained symptom free. The subject relapsed in mid 2006 with an acute episode of psychosis, and was admitted to a psychiatric hospital, where she was commenced on olanzapine 5 mg. Her second (and most recent) admission was in late 2006, in the context of depressive symptoms and command hallucinations. At this time she remained on olanzapine, with the dose increased to 10 mg. There is no reported family history of mental illness, although the patient suspected that her father had suffered from undiagnosed depression, and had self-medicated with alcohol.
Course of trial
An 8 week trial of adjunctive 200 µg transdermal oestradiol was commenced in November 2006, 2 weeks after discharge from the psychiatric inpatient unit. At the baseline interview the subject presented with moderate positive symptoms (PANSS positive = 17), including delusions, grandiosity and hallucinatory behaviour, mild anxiety and depression (associated with amotivation, tension and concentration difficulties; MADRS = 11).
By day 7 an improvement was noted in positive symptoms, with less delusional material expressed and an absence of visual hallucinations. The subject also reported improvements in her concentration. At the completion of the trial (day 56), the positive symptoms had improved considerably (PANSS positive = 10), with no hallucinatory behaviour present, no grandiosity and minimal delusional content. Depressive symptoms also improved with increased motivation and positive feelings about the future (MADRS = 5). There were no significant changes in cognitive functioning, assessed at baseline and day 56.
Discussion
This case highlights the beneficial effects of adjunctive oestradiol as a treatment for psychosis and depression, in women suffering from mental illnesses such as schizophrenia and schizoaffective disorder. This finding is consistent with double-blind studies conducted by our research group [20], [21] and others [35], in which results indicate a significant improvement in psychopathology symptoms in women receiving adjunctive oestradiol, as compared to women receiving placebo. The antipsychotic effects of adjunctive oestradiol may have several explanations. One of the more prominent theories is the role that oestrogen plays in the modulation of dopaminergic [18] and serotonergic [36] neurotransmitter systems, both key neurotransmitters implicated in the efficacy of antipsychotic agents, and thus the pathogenesis of schizophrenia.
Case study 2: SERMs and cognition in a postmenopausal woman with schizophrenia
The subject was a 57-year-old postmenopausal caucasian woman with a diagnosis of schizophrenia. She had responded to an advertisement for a trial examining the use of SERMs in postmenopausal women with schizophrenia. Her medications at trial commencement had been stable for 3 months, and included risperidone (syrup 2 mg per day) and artane (2 mg per day). She was not taking any hormonal therapy.
History
This woman worked as a registered nurse until the age of 30, with the onset of psychosis following the birth of her second child. History included multiple psychiatric hospitalizations, and previous treatment with largactil and modecate. Her most recent hospitalization, in 2005, related to symptoms of paranoia, guarded and stereotyped behaviour, somatic delusions, visual hallucinations and emotional blunting. Her medical history included a hysterectomy, followed by the repair of an umbilical hernia in the 1990s. There was no known family history of mental illness.
Course of trial
A 12 week trial of adjunctive raloxifene hydrochloride 120 mg per day was commenced in August 2006. At the baseline interview the subject was dressed in a bizarre manner, with evident motor retardation. She exhibited minimal positive psychotic symptoms (PANSS positive = 10), while her negative and general psychopathology symptoms were in the mild range (PANSS negative = 20; general psychopathology score = 38). Depressive symptoms were also in the mild range (MADRS = 9).
Over the course of the 12 week trial there was little change in psychopathology, with minimal increase in positive symptoms (week 12 PANSS positive = 12), and only slight improvements observed in negative symptoms (week 12 PANSS negative = 17), general psychopathology (week 12 PANSS general psychopathology = 33) and depression scores (week 12 MADRS = 7); although these variations may reflect normal fluctuations in symptomatology. Improvements were, however, noted in some areas of cognitive functioning (RBANS index scores: Immediate Memory: baseline = 61, week 12 = 61; Visuospatial/Constructional: baseline = 92, week 12 = 72, p < 0.05; Language: baseline = 82, week 12 = 82; Attention: baseline = 88, week 12 = 97; Delayed Memory: baseline = 78, week 12 = 94, p < 0.05). While decreased performance was noted in visuospatial areas, specific improvements were noted in areas of sequential processing and psychomotor speed (RBANS coding subscale; raw scores at baseline = 36 and week 12 = 44), as well as delayed recall of meaningful information (story recall subscale: raw scores at baseline = 5 and week 12 = 8).
Discussion
This case demonstrates the potential for beneficial effects of the SERM, raloxifene, on cognition, specifically with regard to verbal memory and psychomotor speed. Raloxifene at a dose of 120 mg per day has been associated with an improvement in verbal memory (after 1 month of treatment) [38] and a reduced risk of developing mild cognitive impairment in postmenopausal women (following approximately 3 years of treatment) [23].
Case 3: adjunctive Tamoxifen as a treatment for mania
This woman was 44 years old and caucasian, with a diagnosis of schizoaffective disorder, who responded to an advertisement to participate in a trial of adjunctive hormone treatment for women with mania. When she commenced the trial her medications had been stable for approximately 7 months and included lithium (800 mg per day); amisulpride (1200 mg per day); clonazepam (4 mg per day); fluoxetine (40 mg per day) and vitamin D. She had stopped taking the oral contraceptive pill 3 weeks prior to commencing the trial.
Past history
This subject's psychotic disorder began at the age of 31, following the birth of her second child. She was initially diagnosed with postnatal depression, with psychotic features. One year prior to entering the trial, her diagnosis changed to schizoaffective disorder with a differential diagnosis of bipolar affective disorder. History included multiple psychiatric admissions. Her most recent psychiatric hospital admission was in the context of severe depression with psychotic symptoms, 6 months prior to commencing the trial. There was a family history of mental illness, with her sister experiencing depression.
Course of trial
A 28 day trial of adjunctive oral tamoxifen (40 mg per day) was commenced in April 2006. At the baseline interview the subject presented with moderate manic symptoms (CARS-M = 28) including elevated mood, distractibility, increased energy levels, decreased need for sleep, grandiose overvalued ideas and poor judgement. The subject also experienced psychotic symptoms of visual and olfactory hallucinations, with delusional interpretation of these perceptual disturbances (total PANSS = 56; positive symptoms = 18; negative symptoms = 10; general psychopathology symptoms = 28). Moderate depressive symptoms were also evident (MADRS = 20), primarily relating to decreased sleep, decreased appetite and increased lassitude.
As early as week 1 there was a reduction in her manic symptoms (CARS-M = 23). This reduction was further observed by week 4, when her symptoms of mania had significantly reduced to the mild range (CARS-M = 8). Furthermore, the patient's depressive symptoms did not increase as a result of the reduction in her manic symptoms. Interestingly, the patient's psychotic symptomatology remained relatively consistent (day 28: total PANNS = 62; positive symptoms = 14; negative symptoms = 16; general psychopathology symptoms = 32).
Discussion
In line with previous, preliminary investigations [28], [39], [40], the aforedescribed case demonstrates the anti-manic effects of tamoxifen, which are thought to be related to its PKC-inhibitory action [40], [41]. Importantly, the decrease in mania was not accompanied by an increase in depressive symptoms. While the patient's manic symptoms were alleviated, psychotic symptoms remained relatively stable, providing support for research suggesting a pharmacological diversion between mania and psychosis [42].
Conclusion
These three case studies highlight the potential of adjunctive hormones in the treatment of serious mental illness in women. Specifically, adjunctive oestradiol was associated with an improvement in symptoms of psychosis in a premenopausal woman with schizophrenia; adjunctive raloxifene was associated with an improvement in cognitive functioning in a postmenopausal woman with schizophrenia; and adjunctive tamoxifen was associated with an improvement in symptoms of mania in a woman with schizoaffective disorder. While hormonal changes are frequently entwined with psychotic, depressive and cognitive symptoms of mental illness, hormone modulation is not yet common practice in a psychiatric setting. These findings suggest that adjunctive hormone modulation is a promising area of gender-specific treatment for serious mental illness.