On March 27, 2014, the University of North Carolina at Chapel Hill published a press release titled Dr. Cynthia Bulik of UNC leads multinational anorexia genetics project.
Cynthia Bulik, PhD, according to Wikipedia, is the “Distinguished Professor of Eating Disorders in the Department of Psychiatry in the School of Medicine at the University of North Carolina at Chapel Hill, a Professor of Nutrition in the Gillings School of Global Public Health, and Founding Director of the UNC Center of Excellence for Eating Disorders. Dr. Bulik is also professor in the Department of Medical Epidemiology and Biostatistics at the Karolinska Institutet in Stockholm, Sweden and holds an adjunct appointment at the Norwegian Institute of Public Health. She is ranked #2 in the Expertscape list of world experts in eating disorders.”
Here are some quotes from the UNC press release, interspersed with my comments and observations.
“University of North Carolina School of Medicine researcher Cynthia Bulik, PhD, FAED, Distinguished Professor of Eating Disorders, is the lead investigator of the largest and most rigorous genetic initiative in eating disorders ever undertaken.
The Anorexia Nervosa Genetics Initiative (ANGI) will collect clinical information and blood samples from more than 8,000 females and males who have had anorexia nervosa at any point in their lives and those without an eating disorder in an effort to detect genes that contribute to this potentially life-threatening illness.”
In its Spring 2015 update, ANGI announced that it had increased its recruitment goal from 8,000 participants to 13,000.
“‘Genome-wide association studies have been enormously successful in identifying genes that contribute to a range of medical and psychiatric conditions. These discoveries have opened up new avenues of understanding of both cause and cure,’ said Bulik. ‘Once we identify genetic associations in ANGI, we will use the information to develop better strategies to detect, treat, and prevent anorexia nervosa. If our project is successful, it will change the life course of millions of individuals with anorexia and their families.'”
Genome-wide studies may have found some success in general medicine, which deals with real illnesses. But, although I’ve reviewed a good number of such studies in the psychiatric field, I cannot think of even one that has “opened up new avenues of understanding of both cause and cure.” If psychiatry has indeed uncovered new avenues of cause and cure, then it needs to be said that they are being uncharacteristically quiet about them. Throughout psychiatric practice, it’s pretty much business as usual: “diagnose” through facile questionnaires and checklists, and “treat” with neuro-disruptive drugs.
“Each person’s complete set of DNA, or genome, will be purified from a blood sample, placed on tiny chips and scanned on automated laboratory machines. The machines quickly survey each participant’s genome for strategically selected markers of genetic variation, which are called single nucleotide polymorphisms, or SNPs. If certain genetic variations are found to be significantly more frequent in people with anorexia compared with people without anorexia, the variations are said to be ‘associated’ with the disease. The associated genetic variations can serve as powerful pointers to the region of the human genome where the disease-causing problem resides.”
Note the assumption: that individuals who meet psychiatry’s vague criteria for anorexia nervosa have a disease, and the “disease-causing problem” resides in the genome.
“Cynthia Bulik, PhD, director of the UNC Center of Excellence for Eating Disorders, is the lead principal investigator for ANGI, which also includes researchers from the Karolinska Institutet in Stockholm, Sweden, the Queensland Institute of Medical Research in Brisbane, Australia, and Aarhus University in Aarhus, Denmark.”
. . . . . . . . . . . . . . . .
There is additional information about ANGI on UNC’s Center of Excellence for Eating Disorders webpage here.
The first thing one notices is that ANGI has its own very nice logo:
Here are some quotes:
“ANGI represents a global effort to detect genetic variation that contributes to this potentially life-threatening illness. The goal of the research study is to transform our knowledge about the causes of eating disorders to work toward greater understanding and ultimately a cure.”
“If you have suffered from anorexia nervosa at any point in your life, you can help us achieve this goal. If you have never had an eating disorder, but still want to contribute, we invite your participation as well. Your contribution would include a brief 30-minute questionnaire and a blood sample. To make participation easy, we are working with a mobile phlebotomy company that can come to your home, or location of your choice, anywhere in the country, to draw your blood and ship it at back to us at no cost to you.
Only with your participation can we achieve our goal of eliminating this devastating illness.
Participants receive a $25 Amazon gift card.
“We continue to be amazed by the tremendous amount of support we have received from around the country, and by the number of people interested in this groundbreaking study.”
“However, we know that we have not yet reached thousands of individuals who have suffered from AN who may wish to participate in ANGI and donate a blood sample in order to help stop this crushing disorder. If you would like to join the effort to crack the genetic code of AN, don’t hesitate to contact us or take the eligibility survey directly to find out if you’re eligible.”
Your eminent psychiatric researcher needs you!
Of more significance, however, was the fact that the ANGI press release seemed vaguely familiar. So I probed around, and found this:
Since 2007, the University of North Carolina Eating Disorders Program has led a world effort to unite clinicians and researchers around the world in an effort to identify genes that may influence risk for eating disorders. This resulted in the Genetic Consortium for Anorexia Nervosa (GCAN) representing researchers and clinicians from 16 countries around the world. Together with researchers from Kings College London, the UNC program was honored to receive a grant from the Wellcome Trust (WTCCC3) to conduct genomewide association on over 4000 DNA samples from individuals with anorexia nervosa. Members of the consortium gathered information about eating disorders course and genetic material (DNA) from individuals with anorexia nervosa. Results of this global cooperative effort will be released in early 2013.”
GCAN’s final results were published in Molecular Psychiatry, 2014 October; 19(10): 1085-1090, under the title A genome-wide association study of anorexia nervosa. As authors, the article lists 176 individuals plus “The Wellcome Trust Case Control Consortium 3”. Dr. Cynthia Bulik is listed as an author, and also as a member of the Management Committee.
There are four closely typed pages of acknowledgements, which include:
“GlaxoSmithKline (GSK), Leeds (Yorkshire Centre for Eating Disorders). The authors acknowledge the support of the Medical Research Council and GlaxoSmithKline for providing financial support of this project. The support of the Carnegie Trust in the form of a travel award is also acknowledged. We also acknowledge the help and support of the Discovery and Pipeline Genetics, and Translational Medicine and Genetics departments at GSK for their contributions to this study. In particular we would like to acknowledge Mike Stubbins, Julia Perry, Sarah Bujac, David Campbell (at GSK currently or at the time when the study was performed), John Blundell (Leeds University), and Evleen Mann (Yorkshire Centre for Eating Disorders), for their fundamental contribution to the realization of this study.”
“United States. Vanderbilt University School of Medicine, Nashville TN, and the Kartini Clinic for Disordered Eating, Portland, OR. Cases were ascertained from the Kartini Clinic, Portland Oregon. Sample collection and processing was funded by a Bristol-Myers Squibb Freedom to Discover Unrestricted Metabolic Diseases Research Grant to RDC.”
Under the heading Conflicts of Interest it states:
“Conflicts of Interest: Patrick F. Sullivan was on the SAB [Scientific Advisory Board] of Expression Analysis (Durham, NC) [A privately held biotechnology company that specializes in genotyping assays.]. Cynthia Bulik was a consultant for Shire Pharmaceuticals at the time the manuscript was written. Federica Tozzi was full time employee of GSK at the time when the study was performed. David A. Collier was employed by Eli Lilly UK for a portion of the time that this study was performed. James L. Kennedy has received honoraria from Eli Lilly and Roche. Robert D. Levitan has received honorarium from Astra-Zeneca.”
THE GCAN STUDY
Despite the initial enthusiasm, and the truly staggering range of international involvement, the results of the GCAN research were essentially negative: the study did not yield statistically significant results.
Here’s the abstract:
“Anorexia nervosa (AN) is a complex and heritable eating disorder characterized by dangerously low body weight. Neither candidate gene studies nor an initial genome wide association study (GWAS) have yielded significant and replicated results. We performed a GWAS in 2,907 cases with AN from 14 countries (15 sites) and 14,860 ancestrally matched controls as part of the Genetic Consortium for AN (GCAN) and the Wellcome Trust Case Control Consortium 3 (WTCCC3). Individual association analyses were conducted in each stratum and meta-analyzed across all 15 discovery datasets. Seventy-six (72 independent) SNPs were taken forward for in silico (two datasets) or de novo (13 datasets) replication genotyping in 2,677 independent AN cases and 8,629 European ancestry controls along with 458 AN cases and 421 controls from Japan. The final global meta-analysis across discovery and replication datasets comprised 5,551 AN cases and 21,080 controls. AN subtype analyses (1,606 AN restricting; 1,445 AN binge-purge) were performed. No findings reached genome-wide significance. Two intronic variants were suggestively associated: rs9839776 (P=3.01×10-7) in SOX2OT and rs17030795 (P=5.84×10-6) in PPP3CA. Two additional signals were specific to Europeans: rs1523921 (P=5.76×10-6) between CUL3 and FAM124B and rs1886797 (P=8.05×10-6) near SPATA13. Comparing discovery to replication results, 76% of the effects were in the same direction, an observation highly unlikely to be due to chance (P=4×10-6), strongly suggesting that true findings exist but that our sample, the largest yet reported, was underpowered for their detection. The accrual of large genotyped AN case-control samples should be an immediate priority for the field.” [Emphasis added]
This is fairly technical, but the gist is contained in the sentence: “No findings reached genome-wide significance.” In other words, none of the associations between a “diagnosis” of anorexia nervosa and a genetic variant was statistically significant.
There are two additional points in the abstract that warrant some discussion and elaboration. Firstly, the finding that two variants (9839776 and 17030795) were suggestively associated with a “diagnosis” of anorexia nervosa, means that the variants in question were skewed in the direction of a positive association, but were of insufficient magnitude to warrant such a conclusion. In other words, a small number of the “diagnosed” people had the variant in question, but lots of them did not. The skew could simply represent a chance fluctuation in the samples of people tested. The probability values (P = 3.01 x 10-7 and 5.84 x 10-6) seem impressive in the context of most psychosocial and medical research. P = 3.01 x 10-7 means that there is only a 3-in-10-million chance that such a result could be a random fluctuation. In ordinary research, one would accept such a finding with little reservation. But genome wide studies involve literally millions of observations and comparisons, and events with a likelihood of only 3 in 10 million actually occur quite regularly. For this reason, genome wide studies (including the present study) use a significance threshold of P = 5×10-8 (5 in 100 million) to screen for random fluctuations.
Secondly, the reference to discovery-replication comparison refers to the fact that the researchers initially split their participants into two groups. The data from the first group was used to look for associations (discovery), and the data from the second group was used to attempt to replicate any associations discovered. In the discovery phase, the researchers identified 72 variants which were associated weakly (and to a degree that was not statistically significant) with a “diagnosis” of anorexia nervosa. In the replication phase, 55 of these variants showed weak associations in the same direction.
The authors make the point that this suggests that there are some real associations in these variants, but that it will require larger studies to confirm these with a reasonable degree of confidence.
The idea here can be illustrated with a coin-tossing example. Suppose we have a coin that has a 1/500 bias towards heads. For all practical purposes, this is a fair coin and the bias would not be detectable in relatively small trials. But if we toss the coin a million times, we will start to notice a slight preponderance of heads, and if we continue the trials, this preponderance will eventually cross an assigned confidence threshold. But although the difference is real and the analysis is statistically significant, it’s still a very tiny, indeed inconsequential, bias.
In the present context, the authors are essentially making the best of a bad lot. It is reasonable to infer that they had hoped for a more definitive result, and were perhaps disappointed. But they have managed to pull just enough hope from their data to obtain funding for another, bigger, study – the ANGI project mentioned earlier.
Perhaps ANGI will reveal a weak, but statistically significant, association between a gene variant and a “diagnosis” of anorexia nervosa. After all, genes determine the body’s structure, and structure has an affect on our thoughts, actions, feelings, etc. It is likely, for instance, that a large genome-wide study comparing professional cyclists with non-cyclists would show some weak associations with a few variants. This would be of interest to biologists, of course, but it would not establish the notion that professional cycling is a disease, or that it is caused by genes. Nor would it be of much help to cyclists pedaling through the Alpine foothills in the Tour de France.
But even if ANGI reveals a weak, but statistically significant, association between a gene variant and a “diagnosis” of anorexia nervosa, this will not “crack the genetic code of AN”, for the simple reason that no such code exists. What psychiatrists call anorexia nervosa is not a single entity, but rather a loose collection of vaguely defined thoughts, feelings, and behaviors. There is a common core to this aggregate, namely insufficient dietary intake, but the notion that this diverse collection of problems and feelings could be caused by an identifiable genetic code is quite a stretch.
In this regard, it is worth reflecting on the DSM-5’s definition of anorexia nervosa. There are three criteria:
“A. Restriction of energy intake relative to requirements, leading to a significantly low body weight in the context of age, sex, developmental trajectory, and physical health. Significantly low weight is defined as a weight that is less than minimally normal or, for children and adolescents, less than that minimally expected.
B. Intense fear of gaining weight or of becoming fat, or persistent behavior that interferes with weight gain, even though at a significantly low weight.
C. Disturbance in the way in which one’s body weight or shape is experienced, undue influence of body weight or shape on self-evaluation, or persistent lack of recognition of the seriousness of the current low body weight.” (p 338-339)
The first noteworthy feature of this list is that amenorrhea, which had been a requirement in DSM-III and DSM-IV, has been dropped. No explanation for this change is provided. On page 813 of DSM-5 it simply states: “The core diagnostic criteria for anorexia nervosa are conceptually unchanged from DSM-IV with one exception: the requirement for amenorrhea is eliminated.” There is an additional statement on page 341: “Physiological disturbances, including amenorrhea and vital sign abnormalities, are common.” Common, but no required. Again, there’s no explanation as to why amenorrhea is no longer a necessary criterion. It is clear, however, that removing this requirement expands the diagnostic net.
Secondly, it is noteworthy that although DSM-III and DSM-IV had required that body weight be at least 15% below expected weight, DSM-5 requires only that the person’s weight be “significantly low”. The definition of “significantly low”, provided in the DSM-5 (“a weight that is less than minimally normal”) is not a definition at all, and it is clear that this matter is simply being left to psychiatric judgment, opening the door to “diagnosing” people who would not meet the earlier 15% threshold.
Thirdly, DSM-5 follows DSM-III and DSM-IV in listing “intense fear of gaining weight…” as a criterion. But DSM-5 eases the requirement considerably through the use of the following mental gymnastics:
“Individuals with this disorder typically display an intense fear of gaining weight or of becoming fat (Criterion B). This intense fear of becoming fat is usually not alleviated by weight loss. In fact, concern about weight gain may increase even as weight falls. Younger individuals with anorexia nervosa, as well as some adults, may not recognize or acknowledge a fear of weight gain. In the absence of another explanation for the significantly low weight, clinician inference drawn from collateral history, observational data, physical and laboratory findings, or longitudinal course either indicating a fear of weight gain or supporting persistent behaviors that prevent it may be used to establish Criterion B.” (p 340)
In other words, some of the individuals “with this disorder” may not “recognize or acknowledge” their fear of weight gain. In such cases the psychiatrists can decide that they have this fear on their behalf and can thus “establish Criterion B”. Once again, the “diagnostic” net widens.
Fourthly, Criterion C is hopelessly vague and subjective.
Psychiatry routinely justifies these kinds of criteria changes on the grounds that they reflect increased knowledge of the “disease” in question. But this is untenable because there is no definition of the “disease” other than the DSM criteria. There is no disease entity whose pathology has been defined and whose symptoms can be elucidated with increasing precision and clarity as is the case with real diseases. In psychiatry, the only “entity” is the criteria list. This is arbitrary to begin with, and in the absence of any underlying pathology, all additions/amendments are equally arbitrary. This is why psychiatry can continue to widen their “diagnostic” net pretty much at will: there is no reality to which these amendments must conform. “Anorexia nervosa” is whatever psychiatrists say it is.
If we wish to understand what motivates the individuals who systematically under-nourish themselves in these kinds of ways, we need to do two things. Firstly, we need to abandon the empty, disempowering psychiatric labels and their entailed assumptions that one explanation fits all. We need to recognize, that in the area of human thinking, feeling, and behaving, there are always multiple paths to the same place, and that although there will inevitably be a measure of similarity between people’s experiences and motivation, it is through focusing on the uniqueness of each individual that we come to understand his or her perspective. Secondly, we need to sit down with the individual in a spirit of trust and collaboration, and listen to his or her concerns.
It is not a matter of DNA. It is a matter of helping people find some sense of autonomy, purpose, fulfillment, and self-direction in a world that is becoming increasingly isolated, alienated, and divorced from our evolutionary origins.