Mayo Clinic Biobank Begins to Elucidate Bipolar Disorder

Susan Jeffrey

May 07, 2012

May 7, 2012 (Philadelphia, Pennsylvania) — Researchers at the Mayo Clinic aim to provide an important tool for the characterization and investigation of bipolar disorder through development of the Mayo Clinic Bipolar Disorder Biobank.

The project, part of a large general biobank at the institution, aims to provide a resource for clinical, genetic, and epidemiologic study of bipolar disorder, with the goal of individualized medicine in the future.

Manuel E. Fuentes, MD, from the Mayo Clinic, Rochester, Minnesota, discussed some of the baseline clinical and demographic findings from the biobank registry here at the American Psychiatric Association's 2012 Annual Meeting.

Dr. Manuel Fuentes

To date, 708 patients with bipolar disorder, starting with patients with bipolar disorder I and expanding more recently to include those with bipolar disorder II, have been fully analyzed, Dr. Fuentes said. In addition, "Mayo has the possibility to have a really nice control group," he said, noting that the general biobank, including participants without any current illness, now tops 20,000 individuals.

In a separate presentation here, Mohit Chauhan, MD, an Instructor in psychiatry at Mayo Health Systems–Austin, Mayo Clinic College of Medicine in Austin, Minnesota, presented early data on the same patient population but focusing on life stressors associated with bipolar disorder.

By examining the role of these life stressors, Dr. Chauhan said, "perhaps we'll be able to identify better how particular genes interact with a patient's social milieu and impacts the illness."

Genetic Predisposition, Strong Phenotypes

Identification of genetic risk factors associated with the onset of bipolar disorder can potentially lead to early diagnosis and more focused treatment, the researchers note. Bipolar disorder has high heritability, and studies have indicated involvement of a few particular genes, Dr. Fuentes said. Pharmacogenomic predictors have already changed the way some drugs are used clinically, he said.

The Mayo Clinic Bipolar Disorder Biobank is a collaboration between Mayo Clinic, Austin Medical Center – Mayo Health System, the Linder Center of Hope, and the University of Minnesota. The goal is to match the DNA of 1000 bipolar patients to a "rigorous clinical phenotype" for future studies by July of this year, Dr. Fuentes said. After the initial phase, which started in January 2009 and ended in December 2011, the hope is to allow access to the biobank to researchers worldwide. They also intend to begin recruiting patients from other parts of the world.

Patients are identified using 2 research instruments: the Structured Clinical Interview for DSM-IV (SCID), including Module D: Lifetime Mood; and the Bipolar Biobank Clinical Questionnaire, which records information on past illness and treatment, as well as demographics. All study participants underwent venipuncture, with 45 ml of blood drawn from each for DNA extraction and analysis.

Patients were excluded if they were not able to speak English, if they did not provide informed consent, or if they were acutely psychotic or actively suicidal, Dr. Fuentes noted.

Currently, 841 patients have been enrolled, but the analysis presented here focuses on the first 708. The majority (56.5%) are women in their 40s (average age, 44.2 years), he said. More than 90% are white, reflecting the ethnicity of the region from which they are being enrolled. They are highly educated, with 85% having at least some college education. A slim majority, almost 58%, are not working, and almost one half are married. The age of participants is well distributed.

Of the 708 participants, 505 have bipolar I disorder, with the remainder having bipolar II disorder, reflecting the later addition of patients with this latter type of disorder. The most recent episode before enrollment and collection of the blood sample for most was depression, although the severity of the episode was not assessed.

Antidepressant-induced mania was seen in 15.4%, somewhat fewer than has been described previously. "We are assessing very rigorously for this phenotype because it's going to be one of the first analyses we're going to try to do with the genotype," he noted. A history of psychosis was present in 49%, consistent with other samples.

Drug-related adverse events included tardive dyskinesia, seen in 3.5%, associated with a mean length of treatment with antipsychotics, particularly atypical antipsychotics, of 8.4 years. Treatment emergent rash was seen in 7.7%. The rash was life-threatening in 5 patients, or 1.1% of the total sample, but it was life-threatening in nearly 13% of those who had a rash. The reaction was associated with treatment with lamotrigine or carbemazepine.

Family history of bipolar disorder was seen in 55.1% of bipolar patients, major depressive disorder in 81.7%, and completed suicide in 9.9%. This latter finding is slightly higher than seen in other samples, Dr. Fuentes noted. "This could be a marker that we are recruiting a really severe bipolar sample," but it could also represent recall bias. The mean number of mental health diagnoses among members of patients' families was 2.4.

Body Mass Index (BMI) in this sample showed that only 25% were of normal weight or were underweight, leaving 75% overweight or obese. This rate is twice that seen in the general population of the states that most of these patients are from, Minnesota and Ohio, he noted.

However, despite the high rates of obesity and concerns about metabolic issues associated with the medications these patients are receiving, hypertension and cardiovascular comorbidities were seen in only 37%, only slightly higher than 35% seen in the general population of the United States. Information is not yet available on comorbidity with diabetes in their data, he noted.

Finally, in terms of psychiatric comorbidities, the major comorbid condition was generalized anxiety disorder, seen in 41%. Social anxiety was present in 23%; and panic disorder was seen in 23%. There was also a high rate of substance abuse — 25% of participants had nicotine dependence, and 19% had alcohol use disorder.

With these observations, he concluded, "we can say we have a really homogeneous sample, which is a limitation in terms of generalizability," but it may be an advantage for the genetics studies to come.

Life Stressors in Bipolar Disorder

In a separate presentation here, Dr. Chauhan discussed information collected on the history of life stressors reported by these patients. Bipolar disorder is clearly heritable, but not all of those at risk develop the illness. "It is a combination of the genes that we have, the genetic constitution that we have, and also the life stressors we experience," he said, an effect that has been shown clearly, for example, in the depression literature.

Another problem is that genetic constitution may provide a temperament that increases the odds of that individual facing certain life stressors and that also affects how that individual responds to those stressors, he said. Some of the problems may be addressed with projects such as the Mayo Biobank, "if we have bigger numbers, with very rigid phenotype, as rigid as we can, and then look both at their life stressors and their clinical characteristics, as well as the genetic constitution."

For this assessment, patients were presented with 16 negative life stressors, including the following:

  • problems with spouse or significant other

  • lack of family support

  • lack of person the patient can trust and confide in

  • unemployment problems

  • financial problems

  • problems with housing

They were then asked about positive life stressors, including the following:

  • graduating from school/college

  • engagement

  • marriage

  • birth of a child

  • retirement

Patients noted whether these stressors were present or absent, and then rated the severity of the stress associated with this factor on a scale of 0 (no impact) to 3 (severe impact). They were asked first about the presence of these stressors during the 12 months before their first episode of illness, and then during the 12 months before their most recent episode before enrollment.

They found that for total negative life stressors in the 12 months before their first ever episode, about 50% of patients said they had experienced anywhere from 1 to 5 of the negative life events. "So of the 16, a big number of participants were endorsing multiple life events, and this is important because there have been some studies that have shown it's not 1 big traumatic event that is associated with mood episodes but rather a collection of life events — serial life events that collect over time and that lead to either a worse course of illness or mood episodes emerging," Dr. Chauhan noted.

A similar trend was seen for total negative life stressors in the 12 months before their most recent episode, but with even more events reported. "In the past, studies have shown especially in major depression that participants don't need as many life events for future episodes. Their first episode might have happened under serious stressors, but future episodes can happen with minor stressors," he noted. "Our participants are reporting a greater number of life events in the most recent episode prior to study participation."

Comparing negative to positive stressors for first vs most recent episode, they again found that significantly more negative and total events were reported before the most recent episode than before their first episode (P < .001 for both comparisons). There was no difference for positive events, he said.

There was no difference on this parameter with regard to sex, but bipolar II patients reported a greater number of positive life events before the most recent episode, a finding that held after adjustment for age and regardless of whether the most recent episode was hypermania or depression.

As expected, patients with comorbid anxiety, eating disorders, substance abuse or dependence, or attention deficit/hyperactivity disorder reported significantly higher numbers of negative life events than those without such comorbid conditions.

One possibility is that the number of events before the most recent episode is higher because of recall bias, in which more recent events are most easily recalled, Dr. Chauhan speculated. Another is that their questions that related to the first episode were not developmentally staged, "so they do not apply to younger populations, so perhaps they had fewer number of life events to pick from."

Another factor that may affect these results is the issue of "stress priming," meaning people could become more sensitive to stressors as the illness progresses, or the illness itself may set up patients to experience more stressors in their day to day lives.

Their work, he concluded, provides a "plausible model for gene-environment interaction." In future, with study of more severe patients — for example, those with more hospitalizations or more suicide attempts — "perhaps we'll be able to identify better how particular genes interact with patients' social milieu and impacts the illness."

There is currently a lot of interest, Dr. Chauhan concluded, in models that suggest that life stressors may lead to changes in DNA methylation and histone modifications, leading in turn to changes in how the gene is expressed, "and perhaps that leads to more episodes."

The Promise of Biobanks

Peter M. Thompson, MD, professor in the Department of Psychiatry and Director of the Southwest Brain Bank at the University of Texas Health Science Center, San Antonio, moderated the session in which Dr. Chauhan presented his data.

Dr. Thompson acknowledged some bias in his view of biobanks. "I have a brain bank, and I'm also part of the health science center in San Antonio Bipolar Center, which also has a biorepository, so I'm very supportive and very enthusiastic about biobanks in terms of providing tissue for investigators to really have a better handle on identifying markers for treatment response, side effects, and for understanding the pathophysiology of the disease," he told Medscape Medical News.

What the Mayo Clinic is reporting here is "right on track to providing investigators with the tissue they need to better study bipolar disorder."

The studies presented here are important basic work to provide baseline data. "Once you have the tissue, you have to know who the tissue belongs to," Dr. Thompson noted. Careful clinical analysis is needed to determine the participants' phenotype, their levels of depression, and how they are treated. "Do they get manic, do they have thought disorders? You need that information," he said.

Dr. Fuentes, Dr. Chauhan, and Dr. Thompson have disclosed no relevant financial relationships.

The American Psychiatric Association's 2012 Annual Meeting. Abstract #SCR04-1, #SCR05-2. Presented May 5, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.