A Paper About the “Serotonin Hypothesis” Stokes Controversy
Even though depression is one of the most common conditions in the world and we know many of the things that increase the chances of getting it (e.g., early life stress, being female, serious medical illness, grief and loss), scientists still do not understand precisely what causes it. When it comes to treating depression, we have many effective interventions, including a range of evidence-based psychotherapies and antidepressant medications, but once again knowledge about exactly how any of these work to alleviate depression is lacking.
A recent paper and subsequent commentaries that seemed to debunk one theory about how antidepressant medications work received a lot of attention and led its authors to decry the use of these medications. Written by European authors, the paper is a systematic “umbrella” review of the “serotonin theory of depression.” The authors conclude that there is little evidence supporting the serotonin hypothesis and therefore that a whole class of antidepressant drugs, called selective serotonin reuptake inhibitors (SSRIs), may not actually work. In a bold statement the authors of the review paper concluded “Our study shows that [the serotonin theory] is not supported by scientific evidence. It also calls into question the basis for the use of antidepressants.” This is a classic example of authors of a paper making conclusions that go well beyond their data.
The Serotonin Hypothesis of Depression
What is the serotonin hypothesis of depression? Many antidepressant drugs have the property of quickly increasing the amount of a neurotransmitter in the brain called serotonin or 5-HT. Serotonin then binds to specific receptors in the brain, one type of which is also the site where the SSRIs and other types of antidepressants bind. This raised the question decades ago in scientists’ minds of whether low brain levels of serotonin or some decreased ability to utilize serotonin might be a cause of depression. As the paper by Joanna Moncrieff of University College, London and colleagues published in the journal Molecular Psychiatry showed, efforts to measure serotonin in blood, urine, and cerebrospinal fluid of patients with depression have failed to show any deficits, but this is not surprising because the levels of neurotransmitters in the brain are poorly reflected by levels in these peripheral bodily fluids. Moncrieff and colleagues also reviewed studies of various receptors and genes connected to the serotonin system and once again did not find any evidence supporting the simplistic serotonin hypothesis. In the Molecular Psychiatry paper, they concluded “Our comprehensive review of the major strands of research on serotonin shows there is no convincing evidence that depression is associated with, or caused by, lower serotonin concentration or activity.”
So far, Moncrieff and co-authors make a reasonable case that low levels or activity of serotonin is probably not the cause of depression, nor do SSRIs and other antidepressants work simply by increasing the amount of serotonin in the brain. Yet, this conclusion is hardly news: very few scientists actually subscribe to such a simplistic version of the serotonin theory at this point. From the very beginning, even before SSRI antidepressants were first approved in 1987, scientists noted that the drugs increase serotonin levels after just a few doses, but it takes weeks of administering them before depression symptoms actually begin to respond. Obviously, something other than simply low levels of serotonin must be involved. All Moncrieff et. al. did was to accumulate years of research about this in one paper, a useful service perhaps but not a novel observation. Where things get controversial is in the conclusions they draw from their review.
Overreaching from Data to Conclusions
“There is no other accepted pharmacological mechanism for how antidepressants might affect depression,” Moncrieff and co-author Mark A. Horowitz write about in a paper about their research in The Conversation. “If antidepressants exert their effect as placebos, or by numbing emotions, then it is not clear that they do more good than harm…We conclude that it is impossible to say that taking SSRI antidepressants is worthwhile, or even completely safe.”
Does the fact that a theory about the mechanism of action of a medication turns out to be unsupported mean that the medication is ineffective or unsafe? Obviously, this is not automatically the case and the conclusions that Moncrieff and Horowitz reached are themselves unsupported by any data they provide. The FDA approves antidepressants if there is sufficient evidence they are superior to placebo in clinical trials. All the available antidepressants in the U.S. have met that test. That doesn’t mean of course that antidepressants work for every patient with depression, that they are sufficient to completely eliminate depression even in patients for whom they have some benefit, or even that they are the best treatment for depression. Many patients with depression will do better receiving psychotherapy than medication; some need a combination of the two interventions; and some will do best with antidepressants alone. Nor does the fact that the antiquated serotonin hypothesis is not likely to explain how antidepressants work have any relevance to the issue of their safety. Moncrieff and Horowitz seem to believe that antidepressants merely “numb” emotions and are no better than sugar pills, but nothing in their review paper substantiates that claim.
In another commentary, Srijan Sen, head of the University of Michigan’s Depression Center, asks “Do we need to understand exactly how a drug or a non-drug treatment works in order to use it? No — if that were true we would have no treatments for depression, whether it’s drugs, like SSRIs, psychotherapies like cognitive therapy, or lifestyle changes like more consistent sleep patterns.” And Sen notes that “by coincidence” another study was published at around the same time as the Moncrieff paper that does show evidence for involvement of the serotonin system as a cause of depression. That paper, published in the journal Translational Psychiatry, is quite technical and did not garner media attention. Yet it suggests that a combination of adverse life experiences and mutations in a serotonin receptor gene does seem to be involved in why some people become depressed.
Although Moncrieff and co-authors are careful to caution people on antidepressants not to stop them without discussing with their prescribers, their warnings seem half-hearted. Implicit in their comments is the message that if what they misunderstand to be a prominent hypothesis about how antidepressants work—correcting low levels of serotonin in the brain—doesn’t hold up then maybe the use of antidepressant medication is called into question. Media seemed intrigued by that message, which is clearly not a logical conclusion. What they have simply provided us with is another reminder that the human brain is incredibly complex and difficult to study and that in many instances treatments for human diseases, especially when they involve the brain, work without our knowing why they work. The easiest thing to do here is to call for more research, which of course we do and will happen. More to the immediate point, however, is to urge scientists to be cautious in leaping to conclusions and the media to put things in the correct perspective.