Advertisement - Members don't see this ad
-
CycleTrack and SDN are teaming up to make medical school admissions more transparent and accessible! Read the announcement to learn how we’re supporting the future of CycleTrack’s free application-tracking tools .
Big Leap For Schizophrenia!
Started by Blitz2006
Get help with your application
Use all the free resources available to you from SDN: articles, guides, expert advising, forums discussions, and school research.
yawn
Advertisement - Members don't see this ad
I think it's one of those puzzle pieces that helps you finally tie together a bunch of separate sections that you've been stuck on for a long, long time.
Really impressive study. From data mining to mouse model to path/phenotype confirmation. It's rare to see the genetics translated so directly like this. Not to mention they had to come up with some new methods for doing the data mining.
It's incredible. To go back from a genetic risk to postmortem brain tissue to an animal model and find a pathophysiological mechanism and a robust association. I don't think even the optimists would have thought this is possible at this stage in a disorder like schizophrenia.
I think this research is very cool, and hopefully an early step toward understanding the pathophysiology of schizophrenia!
That said (from the article):
"Carrying a gene variant that facilitates aggressive pruning is hardly enough to cause schizophrenia; far too many other factors are at work. Having such a variant, Dr. McCarroll estimates, would increase a person’s risk by about 25 percent over the 1 percent base rate of schizophrenia — that is, to 1.25 percent."
So there is still a great distance to go.
That said (from the article):
"Carrying a gene variant that facilitates aggressive pruning is hardly enough to cause schizophrenia; far too many other factors are at work. Having such a variant, Dr. McCarroll estimates, would increase a person’s risk by about 25 percent over the 1 percent base rate of schizophrenia — that is, to 1.25 percent."
So there is still a great distance to go.
How will this help me treat psychosis better?
Immediately? It won't.
50 years from now? Potentially treating underlying pathophysiology.
If they could connect this research to the role of neurotransmitters in the disorder, then it might be more exciting. The headline stating that they are moving closer to understanding is a bit overstated. That would be like saying that a one yard gain from the 99 yard line means you are closer to a touchdown. Nevertheless, it sounds like a good study and any step forward in neurobiology is a good thing.
But as the article mentions, the risk factor posed by the gene may become much greater and highly relevant if one considers the class of young people who show other signs of developing the disease. It may be that a therapy that blocked the effects of that gene could prevent, say, 20% (or indeed far more -- it may be that that gene is necessary but far from sufficient to developing schizophrenia--it's just not obvious from the description in the article what its prevalence is in the larger population or in schizophrenics) of those now developing schizophrenia from developing it.
Last edited:
So... you're saying we're neurologists now?
If your taking a shot at the idea neurologists can't actually treat anything that seems like an outdated theme from 30 years ago
Edit: maybe thats not what you meant at all and I misinterpreted?
I was surprised this made it into Nature. I'm guessing based on the press coverage they got for this, they'll be looking for a lot more mental health studies.
We have more candidate genes for schizophrenia and autism than we have cable channels, all of which account for a profound minority of actual cases. I guess this is the first "hard" evidence to support the pruning theory, something assumed based on age of onset but with no real biogical support.
That being said, does anyone know if they made an animal model/knockout using this gene?
Do we have the capability of blocking the expression of a gene? A lot of people with genes linked to diseases would appreciate that one.
Advertisement - Members don't see this ad
Not in general of course.
On the other hand, this might be the sort of thing CRISPR/Cas9 might readily do some day.
They did. That's why I said it was such an impressive study.
From the article: "We then evaluated whether mice deficient in C4 had defects in synaptic remodelling, as has been described for C3-deficient mice29. Mice lacking functional C4 exhibited greater overlap between RGC inputs from the two eyes (P < 0.001) than wild-type littermate controls, suggesting reduced synaptic pruning (Fig. 7d, Extended Data Fig. 10d, e and Supplementary Methods). The degree of deficit in C4–/– mice was similar to that previously reported for C1q–/– and C3–/– mice29, 31. Heterozygous C4+/– mice, with one wild-type copy of C4, had an intermediate phenotype (Fig. 7d). These data provide direct evidence that C4 mediates synaptic refinement in the developing brain."
I'm saying that theoretically if this is the pathophysiologic mechanism underlying schizophrenia and a treatment is developed to fix the problem, then you will be able to treat psychosis via primary prevention in the future. If you have to be a neurologist in that case, then I guess so.
Are you saying psychiatry is god of the neurology gaps?
Last edited:
I guess I have to chip in now since I started the thread!
I personally think this study is solid, because it now gives a more scientific/concrete answer to the cause of schizophrenia. Of course, in the clinical world, it doesn't help psychiatrists much. But the take home message from this study is that we are moving into a direction of understanding the cause of this deadly mental illness.
Moreover, mainstream public/medicine likes concrete findings. One of the reasons why psychiatry has lagged behind in "prestige", for lack of better words. But with studies like this, it helps put psychiatry on the map of "cutting edge neuroscience" and our fellow physician colleagues can better understand the disorder because now it involves vocab they understand, like "genes" and "synapses". Perhaps in the next 20 years, the foundations of this paper will become mainstream in describing the causes of schizophrenia. We have been using the dopamine hypothesis for a long time now, we need newer theories and models to help shine a light on this mental illness.
So while this may not add that much to psychiatry clinically, exposure like this only helps our field. Only 2 months ago NY Times published a big article on John Kane/RAISE study.
Now this. Psychiatry is all of a sudden becoming a "hot topic".
Its a good thing,
I personally think this study is solid, because it now gives a more scientific/concrete answer to the cause of schizophrenia. Of course, in the clinical world, it doesn't help psychiatrists much. But the take home message from this study is that we are moving into a direction of understanding the cause of this deadly mental illness.
Moreover, mainstream public/medicine likes concrete findings. One of the reasons why psychiatry has lagged behind in "prestige", for lack of better words. But with studies like this, it helps put psychiatry on the map of "cutting edge neuroscience" and our fellow physician colleagues can better understand the disorder because now it involves vocab they understand, like "genes" and "synapses". Perhaps in the next 20 years, the foundations of this paper will become mainstream in describing the causes of schizophrenia. We have been using the dopamine hypothesis for a long time now, we need newer theories and models to help shine a light on this mental illness.
So while this may not add that much to psychiatry clinically, exposure like this only helps our field. Only 2 months ago NY Times published a big article on John Kane/RAISE study.
Now this. Psychiatry is all of a sudden becoming a "hot topic".
Its a good thing,
Last edited:
This is kind of a random question, but is what they discovered at all related to the current understanding of Alzheimer's?
The neurology statement is that we're adopting a pattern here that basic sciences is making leaps forward without any practical use or treatment options. We need to temper excitement and continue the trudge forward with the limited care options.
Interesting. Thank you for the detailed reply. I have lots of follow-up questions. But I'll just pick one: What do you mean when you say that schizophrenia doesn't exist in nature--is that to say it doesn't exist outside of humans?
you can read more about the idea of natural kinds - here, here, here, and here
Psychiatric disorders in our current taxonomy "carve nature at the joints" assuming there is a difference between normal phenomena and mental disorder, and that different diagnoses represent different kinds of disorders that map out to nature. Elsewhere Nassir Ghaemi has argued that psychiatric diagnoses are "ideal types" in the Weberian sense. The problem with this is that much of our research is essentialist - assuming that these arbitrary categories map onto biological phenomena, when they don't. This has been one of the challenges and limitations for psychiatric genetics. How can you expect to find the genes "for" something that doesn't even exist? To put it another way, Alzheimer's is disease with a very specific pathology, but "schizophrenia" is not. Rather than being akin to Alzheimer's it is more like "cognitive disorder". Cognitive disorder isn't a disease of course - there are lots of different kinds of memory loss (long-term, short-term, complex attention, learning, perceptual-motor, social cognition) and there are lots of different causes. "Schizophrenia" is both heterogeneous in concept (such that there are a bunch of different presentations which are labeled as schizophrenia that don't have much in common) and in etiology.
I know you asked Splik, but here's my probably less informed take -- throughout both Western and Eastern history, there have been documented cases of severe mental illness and intellectual disability. These people were "taken care of" by the community, which ranged from being locked in the basement to maybe even their own house/hut. With industrialization and urbanization, that was no longer a viable system (and, less cynically, the Enlightenment pushed people to recognize individual human liberty). Thus, the asylum is born, and huge swaths of the population that are not able to be take care of themselves (for whatever reason) are clustered together. We start breaking people into those who were born disabled and those who developed their symptoms later in life. Then we divide the latter again, those who get better and those whose symptoms wax and wane. The former become schizophrenia spectrum whatever, the latter become bipolar spectrum-whatever. The labels are helpful as a society figuring out what to do with these people, but they're still artificial constructs.
A similar process happened with autism. You had several kids in our educational system who needed help, although didn't fit any formal diagnosis. The definition of autism was loosened and services were provided and all of a sudden you have a massive increase in diagnoses. Do all of these kids suffer from the same pathophysiology? Probably not. We just sort of lumped them together.
Alzheimer's is different. It was "discovered" in the turn of the century, when human lifespan was much shorter and dementia was much more rare -- the first case described was a 50 year-old woman. Instead of taking a huge swath of the population and sticking a label on them, this was a relatively small cohort that could be identified and studied at an almost individual level (a bottom-up approach instead of a top-down). I think that's changing now as people age -- I'm sure there are a ton of people diagnosed with Alzheimer's that likely have a bad combination of vascular dementia, corticobasal degeneration, non-Alzheimer's tauopathy and even pseudodementia, and its becoming an artificial construct of its own.
That's great, I'll definitely check it out when I have access. Should be interesting to see how the animal model progresses -- I'm curious when this synaptic remodelling occurs in the wild-type vs. KO. It seems like the model would only be applicable if the pruning occurs late in life.
I still feel this has been sensationalized, given the hundreds of candidate genes we have for schizophrenia that accounts for a fraction of the people who actually have the diagnosis. But I'm glad psychiatry is getting some good press.
The one thing I would add to this is that within this artificial construct of schizophrenia there probably are some real pathological processes or distinct mechanisms of dysfunction. It is difficult to research though when you start with a mistaken premise.
To use the dementia analogy, if we held onto the perspective that there was one dementia and didn't separate out the different causes, the only predictor would be age. When we group by only symptoms, completely unrelated diseases get lumps together.
Autism is another example. The defining characteristic is social impairment. Talk about non-specific and the potential for numerous causes. We now realize that early intervention can dramatically affect the outcome of children with this "disorder". My niece's child met all of the diagnostic criteria for this disorder and was delayed in multiple milestones. After intervention by grandparents, speech therapist, occupational therapist, and psychologists she was mainstreamed by 3rd grade and is now functioning above average in most measures although she is still a bit geeky and compulsive. Without intervention, she was not developing language and was withdrawn, self-injurious, and completely dysfunctional. Maybe she wasn't autistic, but she had all the signs and symptoms and since we don't really know what autism is other than a broad category with a few broad symptoms that are non-specific. Who really knows? Her family expected her to get better and she surpassed all expectations. When we as society don't expect people to improve, they don't.
I'm pretty sure it's open access.
http://www.nature.com/nature/journal/vaop/ncurrent/full/nature16549.html
H
HarryMTieboutMD
This is an AMAZING paper, truly Kraepelinian.... and please, read the actual paper, not just the NY Times article. The methodology is sophisticated and the experiments are most elegant, which, along with the novelty of the subject is why this is published in Nature. I think papers like this are vital for psychiatrists to read (and I haven't been in a wet lab since undergrad), but any deeper understanding into the underlying neurobiology, however rudimentary it may be, definitely changes and even enhances the way I interact with patients and their families. Is "increased C4A expression leading to increased C3/compliment casacade leading to increased synaptic pruning leading to cortical thinning leading to Dementia Praecox" completely explanatory? Of course not, and the authors are not arguing that it is. Does this provide more concrete evidence that will prompt future thinking and rigorous research in attempts to expand upon/contradict/clarify the issues presented... most definitely. The chair of my department is one of the most formidable basic scientists in all of psychiatry and has done great volumes of work on the glutamate and GABA systems. He once told me, "you know, there is a great possibility that everything biological psychiatry is doing is completely wrong... but the important thing is that the effort is there because that's from where progress starts" (rather than just throwing in the white flag and saying OK let's just stick to Cartesian Dualism)
While yes, ultimately we still treat psychosis by dropping the D2 antagonist (or Clozapine) bomb and just deal with the collateral damage, I think the more informed one is about suggested mechanisms for underlying pathophysiology, the better the clinician he or she will be.
While yes, ultimately we still treat psychosis by dropping the D2 antagonist (or Clozapine) bomb and just deal with the collateral damage, I think the more informed one is about suggested mechanisms for underlying pathophysiology, the better the clinician he or she will be.
Last edited by a moderator:
Maybe its amazing... honestly, most of the methods section is way above my head. My initial reaction was to the news (and not the article itself). In my mind, its equivalent to the headline "Scientists discover the genetic secret to intelligence!" Obviously, there's some hereditary component to intelligence, and I'm sure we could put together a ton of candidate genes to figure out the mechanism for why SOME people are smart. Would that explain every intelligent person on the planet? No. Would it still be a worthwhile endeavor? Sure. I'm not a complete nihilist, but just wanted to add some perspective.
As for the Nature publication -- a thoughtful, elegant study is necessary but not sufficient to be published in a high impact journal. The editors are very attuned to what is trending and what can attract media attention (hence the barrage of crappy SSRI articles in BMJ), which is why most people heard about this from the mass media as opposed to didactics.
And no, this paper is not open-access, which is frustrating in and of itself. I already have a gripe that publically-funded studies are not available to the public. I now think we need an additional law that any article given a major press release should also be open-access.
Nope. I ended up getting it at work, but I'm pretty sure the only Nature papers that are open are Nature Communications.
H
HarryMTieboutMD
Yeah, "novelty of the subject" implies "this is a hot topic that will turn heads." And the intelligence argument is not a good comparison for a multitude of reasons. First, assuming you are speaking of IQ (and there are other measures of intelligence, Ken Heilman has written extensively about creative intelligence), it is continuous metric though is not homogenous across it's distribution (ie, at some point, there is a definite pathological state). The criteria for schizophrenia starting from DSM III (and has not changed that much since) represent an entity in which there is a clear boundary between "sick" and "not sick" (this is not any different than any other field in medicine, absent a biomarker). So, when you have thousands of subjects (who are also patients) diagnosed using very similar criteria, you are studying a somewhat consistent clinical phenomenon. Danny Weinberger has brilliantly articulated editorial related to the subject recently published in JAMA Psych: http://archpsyc.jamanetwork.com/article.aspx?articleid=2469109
Advertisement - Members don't see this ad
thanks for the link - interesting article. Though I have to say the criteria for schizophrenia have changed significantly from DSM-III becoming increasingly restrictive and narrow. DSM-III schizophrenia were essentially Schneiderian, and until DSM-5, FRS were given special significance in the schizophrenia criteria. also bizarre delusions were given special significance until DSM-5. Finally, the Kraepelinian subtypes of schizophrenia (catatonic, paranoid, hebephrenic/disorganized) have been expunged from the current classification. Many cases that would previously have been "schizophrenia" would now be delusional disorder or an "other specified psychotic disorder". The DSM criteria have become increasingly divergent from the ICD-10 which is still Schneiderian (and also only requires 1 month of symptoms thus includes DSM "schizophreniform disorder")
The problem with "sick" vs "not sick" in DSM is this is not defined by symptomatology or any kind of phenotype and instead defined by "distress" and functional impairment which are not genetically determined but socially determined. Whether someone is impaired by their symptoms has everything to do with the social environment, attitudes, support etc, and nothing to do with genetics or the neurobiology of the illness. Though it has been clinically necessary to distinguish those who are impaired by their psychosis from those who are not, this doesn't help us to understand the neurogenetic factors that may lead people to develop delusions and hallucinations in the first place.
I agree re: IQ, which is why I deliberately said intelligence. I think what I'm saying is consistent with the article you cite -- someone can be short for several different reasons, and while we might find a gene that confers shortness, that would not be sufficient to explain everyone with a low height.
As for the DSM, if there is a discrete cut-off for when someone is "sick" and "not sick", I'm having a hard time finding it. Just looking at Criteria B for schizophrenia, and there's more hand-waving in one sentence than at an entire conductor school (sorry, that was bad...). We might end up finding people with this MHC/C4 variant WITH psychosis, and they could still potentially not be diagnosed with schizophrenia, provided they are functioning in society.
Like I said, I'm not a total nihilist, but I think its good to have some perspective. In the early 2000's, there was a lot of excitement over the serotonin transport gene (http://science.sciencemag.org/content/301/5631/386). Presumably, people with the short allele responded to stress differently and developed depression. Over the span of a decade, the reality has been much more complicated, and some studies have straight out debunked it as an explanatory model.
This is different though. It's not simply finding a gene as a risk factor. We already found a lot of them. They went further and correlated gene expression with risk and found a very nice correlation then made a mouse model and found a compelling pathophysiological mechanism that explains some of the other evidence. It's new and groundbreaking for all of these reasons because there's more of an anchor here that we could target in future research. The innovation is mostly on the genetic side because the locus was found before but it seemed that the genetics were too complex to get down in the nitty gritty and find the alleles. And that's why this came from hardcore geneticists rather than neuroscientists. Having said that, the gene is obviously only a risk factor and a small one at that. We need more studies to determine clinical relevance but also basic science to see if there are different loci/epigentic mechanisms leading to the same end result (defect in synaptic pruning) or if it's more complicated than that. It also could be more strongly related to a subtype of schizophrenia, hence the importance of trying to find new classification schemes that depend a little less on phenomenology.
Last edited:
H
HarryMTieboutMD
III-R criteria narrowed the criteria, I believe, and since then the changes have been more or less nuances (though I'm not a fan of V by any means). And I should rephrase "sick" vs "not sick" as having an illness vs not having an illness. Of course the "inhibition in level of functioning" is not strictly a medical judgment, but this is no different than any other area in medicine except that we don't have biomarkers. (eg, someone with an HbA1c of 9 refuses to do lifestyle modification or start meds yet never develops retinopathy, nephropathy, neuropathy, etc and lives a reasonably healthy life... he or she still has diabetes). The point is that we can group these people, who have hallucinations/delusions/disorganization in speech/behaviour/negative symptoms >6 months, some of whom might show little impairment though realizing that there is some underlying pathology and use them for further study. Again this is not different from any other area of medicine.
And completely agree with Jorge's point about how this is truly a different study and how novel and exciting it is. Bear in mind how powerful it is that the proposed pathologic mechanism corresponds with the negative and cognitive symptoms- not hallucinations or delusions- which is why this is work of which Kraepelin would be proud. Obviously this is just the beginning, but I am very excited to see what the best basic scientists in our field do to build on this.
And using the 5HTTP controversy and probable lack of association between allele of transporter, life events, and depression as an argument for the limitations (some people say fallibility) of GWAS in psychiatric disorders is so 2010. That said, I have far less of a problem with GWAS of schizophrenia using our current diagnostic system than I do with GWAS for PTSD...
