The estrogen hypothesis of schizophrenia holds that estrogen has a protective, mitigating effect on schizophrenia in women, and that this helps explain why women tend to present with first episodes later in life, present less frequently than men, and, some studies have found, have better prognosis and treatment response.

A growing scientific and clinical literature on estrogen and the brain has added heft to what was once a contentious theory. Estrogen has been found to modulate dopamine and serotonin transmission. Estrogen levels in women with schizophrenia have been shown to be lower than those seen in healthy women, with illness onset or relapses most often occurring during the phase of the menstrual cycle when estrogen drops. Women also see a higher rate of late life–onset schizophrenia than do men, which might be related to estrogen decreases at menopause (Schizophr. Res. Treatment 2012 [doi:10.1155/2012/916198]).

A handful of studies has explored whether estrogen replacement therapy, in the form of oral or transdermal estrogen, is helpful in patients with schizophrenia. Recent research has begun to investigate whether agents with estrogenlike effects, such as the selective estrogen receptor modulator (SERM) raloxifene, also could be helpful when added to conventional treatment with antipsychotics. Raloxifene, manufactured by Eli Lilly & Co. as Evista, is licensed as a preventive therapy for osteoporosis, and – unlike estrogens – does not affect breast or uterine tissue.

Dr. Jayashri Kulkarni

At the fourth International Congress of Medicine and Women’s Mental Health, held in April in Medellin, Colombia, schizophrenia researcher Dr. Judith Usall of Parc Sanitari Sant Joan de Déu, in Barcelona, presented her ongoing work exploring the estrogen hypothesis and raloxifene. Dr. Usall is one of a small group of researchers working with raloxifene in schizophrenia patients; another leader in the field, Dr. Jayashri Kulkarni of Australia’s Alfred Hospital in Melbourne and Monash University in Clayton, Victoria, previously had tested raloxifene in a pilot study – a small, randomized, controlled trial enrolling postmenopausal women with schizophrenia (n = 26) – and found it to be effective (Psychoneuroendocrinology 2010;335:1142-7).

Last year, in a randomized, placebo-controlled trial enrolling 33 postmenopausal women with schizophrenia, Dr. Usall and colleagues found that adding 60 mg daily of raloxifene – the standard dosage indicated for osteoporosis prevention – to regular antipsychotic treatment improved psychotic symptoms at 12 weeks, compared with antipsychotics alone, with reduction in negative (P = .044), positive (P = .031), and general psychopathological (P = .045) symptoms, and without significant adverse effects (J. Clin. Psychiatry 2011;72:1552-7).

A new trial of similar design, also led by Dr. Usall, began enrolling in January, and will recruit a cohort of 80 postmenopausal women with schizophrenia to determine whether they see a benefit after 6 months of adjunctive treatment with 60 mg of raloxifene.

“My main objective is to improve the treatment of patients with schizophrenia, as well as to improve our knowledge of the etiology of schizophrenia,” Dr. Udall said in an interview. “If our trial and others confirm and expand upon our positive results, I think that the use of raloxifene could be recommended in postmenopausal patients.”

The Australian researchers are taking a somewhat bolder approach, using higher doses of raloxifene and recruiting younger women and men as well as postmenopausal women to their randomized, controlled trials. Dr. Kulkarni and colleagues are close to wrapping up a larger trial in postmenopausal women with schizophrenia (n = 180) using 120 mg daily of raloxifene in the intervention arm. Their pilot study of raloxifene in postmenopausal women had used the lower dose, but “while the 60-mg [dose] gave some cognitive improvement, it didn’t really touch the psychotic symptoms,” Dr. Kulkarni said in an interview.

Dr. Usall said she thought that raloxifene also might be useful in premenopausal women with schizophrenia, particularly those with symptom exacerbations coinciding with their menstrual cycles, but noted that safety information was lacking for both the higher dose and for younger women. “My trial addresses the possible efficacy of raloxifene only in postmenopausal women with schizophrenia,” Dr. Usall said.

The Barcelona researchers also are focusing on the negative symptoms of schizophrenia, setting their primary clinical end point at a 20% or more improvement in negative symptoms. Their reasoning, Dr. Usall said, is that “antipsychotics are not as effective in negative symptoms as in positive ones, and because there are different neurobiological and clinical data that point to the possible efficacy of estrogen in negative symptoms.”

Dr. Kulkarni said that although no formal analysis has yet been conducted for her team’s raloxifene trial in younger women (n = 120), “it’s already looking even better than in the older women. We’re getting positive symptom response plus cognitive improvement.”

Dr. Kulkarni also said that she has seen “dramatic turnarounds” in some postmenopausal patients and among women with whose psychosis onset was post partum, in the intervention arms of her studies. But there have also been slow responders and nonresponders, she said. Predictors of response “will hopefully get clearer as our sample sizes increase. We think there may be an inherited sensitivity to the hormonal milieu.”

The Australian team also is conducting a small trial of raloxifene in men with schizophrenia, designed to enroll 30. Dr. Kulkarni previously had worked with estradiol in men with schizophrenia (Schizophr. Res. 2011;125:278-83), but found that its feminizing effects necessitated that the trial be stopped. The investigators hope that raloxifene, which is more selective, will deliver improvement without the feminizing effects.

Raloxifene is a long way from being a standard of care for schizophrenia in patients of any age or sex, but it is increasingly being used as an adjunct treatment for postmenopausal women with schizophrenia in Australia, Dr. Kulkarni said. Some researchers have begun to recommend it as a supplementary treatment for younger women, too.

In a recent literature review on treating hormone-linked disease exacerbations among women with schizophrenia, Dr. Mary V. Seeman of the University of Toronto concluded that raloxifene could be considered among treatment options for premenstrual symptom aggravation (Acta. Psychiatr. Scand. 2012;125:363-71).

Dr. Seeman said in an e-mail interview that although SERMs for schizophrenic symptoms have been used mainly in research studies and not in general practice, “individual clinicians may prescribe them for individual patients.”

Dr. Kulkarni said that awareness is growing of the potential to add estrogens or SERMs when standard treatment is not enough.

“We’re in a transition phase at the moment, because we’ve gone from [SERMs’] being used purely in research to their being a product that’s out there to be tried; it’s a question of getting the information to practicing clinicians.” A 2012 paper by the Stanley Research Foundation, which has helped fund Dr. Kulkarni’s and Dr. Usall’s clinical trials, recommended several adjunct treatments for schizophrenia, among which were estrogen and raloxifene.

Still, “psychiatry tends to operate in a silo,” Dr. Kulkarni said. “Most psychiatrists are happy to prescribe antidepressants or antipsychotics but may not want to prescribe a hormone because it opens up a whole new clinical area. I think that’s to the profession’s detriment.” Dr. Kulkarni added that as the results of more ongoing trials come in, “we can begin to think about creating clinical guidelines” for adjunctive treatment with SERMs.

Dr. Usall commented that “unfortunately, the issue of gender-sensitive mental health is not sufficiently introduced into clinical practice.” However, she said, clinicians can nonetheless incorporate aspects of the estrogen hypothesis into the management of female schizophrenia patients even without prescribing SERMs.

Exacerbations of disease activity often will coincide with menstrual changes, Dr. Usall said, and symptoms should “always be evaluated with regard to the menstrual cycle.” Clinicians might choose to add estrogen/progesterone combinations, or increase dosage of antipsychotic drugs during the luteal phase, she said.

Dr. Kulkarni said she felt that raloxifene represented “a whole new avenue” of adjunctive treatment, with many drugs like it currently in the pipeline. Dr. Seeman, meanwhile, cautioned that SERMs might represent only the first stage of a new treatment paradigm.

“While SERMs like raloxifene are thought to be safe for breast and uterus, they still have difficulty in crossing the blood-brain barrier,” she said. “I suspect that, with time, safe estrogenlike compounds that enter the brain more efficiently will be developed – maybe they already exist – and will be useful for neuropsychiatric disease, including schizophrenia. It’s still early days but very promising.”

Dr. Usall, Dr. Seeman, and Dr. Kulkarni declared no conflicts of interest related to their research. Dr. Seeman is medical adviser to Clera Inc.