Why Can’t We Find an Antidepressant That Really Works? – Valerie Pavilonis

Lowry Smith, a 57-year-old living in Buffalo, is far from alone: Of people who take antidepressants, many won’t respond to the first drug they try, and some will embark on a multi-month or multi-year quest to find what, if any, drug works for them. But for others, the reality is far less complicated: Catherine Livingston, a 30-year-old mother in Kansas, told The Dispatch that, after starting a low dose of Lexapro about three years ago, she felt “so much better,” after just two weeks—and still takes the medication.

These experiences illustrate the asterisk that’s long been attached to antidepressants: a miasma of uncertainty around which drugs work for which people. Decades after the first antidepressants were discovered, researchers are still seeking to understand what depression is in the first place—and how drugs, from Prozac to the newcomer ketamine, might help the millions of people who suffer under depression’s weight. 

The origins of modern antidepressant medication can be traced back to the 1950s, when Nathan S. Kline, the director of research at Rockland Hospital in New York heard that the tuberculosis drug iproniazid was anecdotally associated with positive effects on mood. After conducting his own research, Kline—along with the psychoanalyst Mortimer Ostow—found that iproniazid was a “psychic energizer” capable of helping some chronically depressed patients. “I saw a fairly well-known young artist who had been unable to produce any new canvases for a period of over a year,” Kline said in 1957, according to a New York Times report. “Treatment with [iproniazid] seemed to ‘break the dam’ and during the summer he produced a profusion of oils, water-colors and sketches totaling more than a hundred.”

Still, Kline’s research came with caveats. Iproniazid came with, among other side effects, a strong risk of liver toxicity. There was also the stark fact that iproniazid worked for just a fraction of the population studied: “[N]ot all the patients were improved, and in some cases the improvement came very slowly,” the Times reported. Psychiatric medicine’s variance was there from the beginning. 

Also there from the beginning was the idea of serotonin—a neurotransmitter and hormone—as playing some role in mood disorders. “The administration [of] iproniazid, it was observed, led to an increase of serotonin in the brain indicating the improvement in the patient’s mood was probably the result of checking the action of the serotonin-inhibiting enzyme,” the Times reported in 1957. “This, it is believed, may open a new approach to the chemical treatment of mental disease.”

Around the same time as Kline’s work, Swiss scientist Roland Kuhn was working at a hospital in a remote village when he discovered that the compound G22355, produced by the chemical company Geigy, seemed to alleviate depressive symptoms in schizophrenic patients. Supported by Geigy, Kuhn undertook a formal study of G22355 in more than 500 patients with various mood disorders and reported in 1957 that many saw their depressive symptoms greatly improve. “The patients get up in the morning of their own accord, they speak louder and more rapidly, their facial expression becomes more vivacious,” Kuhn wrote. “They commence some activity on their own, again seeking contact with other people, they begin to entertain themselves, take part in games, become more cheerful and are once again able to laugh.” G22355, now known as imipramine, was introduced commercially in Europe in 1958 and in the United States in 1959.

Kuhn’s article did not mention serotonin. But later, researchers did indeed find the mechanism by which imipramine seemed to work: by leading to an increase of, among other things, serotonin—just like iproniazid. 

These twin discoveries—and the central role that serotonin appeared to play in them—eventually led to something called the monoamine deficiency hypothesis, which posits that mood disorders stem from a lack of some sort of neurotransmitter, most commonly dopamine, norepinephrine, and serotonin. It’s largely from this hypothesis that drug companies have continued to synthesize new drugs for depression, up to the present-day first-line treatments known as selective serotonin reuptake inhibitors (SSRIs) that increase serotonin levels in the brain. As of 2023, the majority of the 11.4 percent of American adults taking antidepressants are taking SSRIs, including brand-name drugs like Zoloft and Prozac.

Since about 30 million Americans—roughly the population of Saudi Arabia—are taking antidepressants daily, we have a pretty good idea about how they work, right?

Wrong. Despite the changes in serotonin levels wrought by imipramine and iproniazid—and by today’s more modern drugs—it’s unclear whether serotonin levels by themselves are the primary driver of depression. And that’s a problem when it comes to treatment. 

Just as Kline found that some patients were not helped by iproniazid, with current antidepressants, some patients improve while others don’t. A meta-analysis from June 2022 of 232 studies conducted between 1979 and 2016, found that, across 73,388 participants with major depressive disorder, patients given medications improved only slightly more than patients given placebos.

One major trial, starting in 2000, found that only about 30 percent of patients treated with the SSRI citalopram reached remission. Some patients switched to other medications, which slightly increased overall remission rates. But overall, efficacy was low. 

Scientists “don’t really know what’s going on in depression,” Ben Samuels, an associate professor of psychiatry at Rutgers University, told The Dispatch. And our best understanding of the condition, he said, comes from reverse-engineering the drugs that do help some people (which is what led to the monoamine deficiency hypothesis in the first place). 

“The reason that has become a hypothesis of depression is just because a lot of the drugs increase the levels of these neurotransmitters,” Samuels said.

Besides not being effective for large percentages of people, it has been impossible to determine for whom the medications will work before prescribing them. Adding to the complexity of treatment is the range of antidepressants on the market. In addition to SSRIs, other antidepressant options include serotonin-noradrenaline reuptake inhibitors (SNRIs), norepinephrine and dopamine reuptake inhibitors (NDRIs), and monoamine oxidase inhibitors (MAOIs).

Still, evidence of one medication having totally different effects from another—despite the medications’ chemical differences—is scant. Rather, psychiatrists will often prescribe medications based more on family history, side effects, or the potential for drug interactions, and if one medication doesn’t seem to work, a psychiatrist will often start a patient on another. (“There is little empirical evidence to guide clinicians in choosing among the diverse array of antidepressants available,” one group of scientists wrote in 2004.) “There isn’t really any increased likelihood of one working over the other,” Samuels said. “The main differences are in side effect profiles. In my honest opinion, it starts to get a little anecdotal with some psychiatrists where they’ve had a bunch of success with one drug relative to another. So they just kind of stick with it.” 

Jordan Smoller, a psychiatrist at Harvard University and the director of the Psychiatric and Neurodevelopmental Genetics Unit at Massachusetts General Hospital, echoed Samuels’ sentiments. “If you’re starting an SSRI, often the considerations are somewhat nuanced, and there’s a little bit of mythology about which one might be most appropriate to choose, having to do with their differences in side effects, and also how they’re metabolized,” Smoller said. “Some of them are more likely to interact with other medicines than others.”

Part of the seeming randomness in prescribing antidepressants comes from the difficulty in understanding how depression works. Simply put, there is no depression gene. 

Compounding the confusion is the fact that depression, like the vast majority of human maladies, likely has both genetic and environmental components. According to Smoller, a myriad of factors can influence depression and one’s experience of it, from trauma exposure to social support to physical activity. 

Still, while the exact causes of depression are murky, researchers tend to focus on what they can see and measure in order to direct new treatments. Samuel Wilkinson, a professor of psychiatry at Yale University and the associate director of the Yale Depression Research Program, noted that some people with depression show fewer synaptic connections in brain scans (which is why the anesthetic ketamine—which seems to increase those connections—is thought of as a possible treatment). But the brain scans of other people with depressive symptoms can look perfectly normal. 

A look into the Diagnostic and Statistical Manual of Mental Disorders (DSM), a taxonomy of psychiatric conditions published by the American Psychiatric Association, shows the variability of depressive symptoms. According to the most recent edition of the DSM, a depressed person is someone who has experienced apathy or loss of interest for at least two weeks, along with symptoms such as weight/appetite changes, sleep disturbances, slowness of movement, fatigue, feelings worthlessness or guilt, executive dysfunction, and suicidal ideation. The diagnostic problem, though, is that one apathetic person could have, say, fatigue, slow movement, weight gain, and sleep disturbances, while a second person could have a low mood, weight loss, agitation, executive dysfunction, and suicidal ideation. Simply put, two people diagnosed with the same malady could present completely differently. 

Several of the neuroscientists who spoke with The Dispatch noted the complexity in diagnosing depression as a disorder in the first place. “We call this depression, but it’s probably a handful of different things, [a] different kind of collection of features,” Wilkinson said. Lisa Monteggia, a neuroscientist at Vanderbilt University, also noted that there are various types of the condition, from major depressive disorder to postpartum depression. 

A simpler way of understanding depression has to do with the aforementioned monoamine deficiency hypothesis, also known as the “chemical imbalance theory”: In this view, depression and other mood disorders mostly result from an imbalance of neurotransmitters, including serotonin. But the fact that serotonin-increasing drugs don’t work in every depressed person means the hypothesis only tells part of the story. “There may be some truth to it, but it’s clearly not the whole picture, at least not in most patients,” Wilkinson said. 

While “chemical imbalance” used to be a buzzword some decades ago, a newer buzzword in psychiatry is “neuroplasticity”—that is, the brain’s ability to adapt to changing cues in the environment. In brain scans, depressed people sometimes have less dense clusters of synapses.  “We think that’s linked to a reduction in plasticity,” Wilkinson said. 

Enter ketamine, a general anesthetic more commonly known for its hallucinogenic properties and its use by certain tech moguls. Typically, patients will try multiple antidepressants with little success before moving onto ketamine, which comes in the form of both infusions and an FDA-approved nasal spray.

“We’re thinking of ketamine as a therapy not just for depression but for what we call, in the field, treatment-resistant depression, which is when people try Prozac or Celexa or whatever, and they’re still stuck in the state that they don’t want to be,” Wilkinson, who studies ketamine, said.

And it tends to work fast. “People typically respond within a couple hours if they’re going to respond,” Monteggia said. That poses major questions about how older depression treatments work. SSRIs and similar drugs typically take weeks to have a therapeutic effect, if they’re going to have one at all, and the theory behind this lag is generally that some sort of repair work is occurring within the brain. But ketamine’s speed indicates that depression may not be caused by broken structures or processes at all.

“The puzzle is not done by any means, but there are a number of pieces that are being filled in,” Wilkinson said. “This is, we hope, we think, [a] more accurate picture than the chemical imbalance thing that we spoke of 20, 30 years ago.” 

A danger with ketamine is that the drug can cause dissociation and sometimes hallucinations—at high doses, it can be used recreationally to get high—so it’s safest to try the drug in a controlled clinical setting. Still, ketamine has been helpful for some, even if it’s not effective for everyone. Indeed, Lowry Smith credits ketamine with helping her out of depression: “I felt so much better,” she told The Dispatch, recounting her first ketamine infusion. “The next day I posted on Facebook a picture of a ballerina doing, I think it’s called arabesque, where they jump, one leg is forward and the other leg is back. That was my expression of how I felt the next day.” Shortly after she started ketamine, she stopped taking antidepressants completely. 

Still, researchers haven’t completely moved on from SSRIs. In an effort to streamline depression treatment, some researchers are seeking to understand more about how—and for whom—SSRIs actually do work. 

Mark Rasenick, a professor of physiology and biophysics at the University of Illinois at Chicago, pointed out that a major obstacle in treating depression comes from the difficulty in diagnosing a largely intangible problem. So he and his team have developed a blood test that measures cellular-level reactions typical in depressed people. The blood test has a high degree of accuracy, over 90 percent, but false positives and a smaller sample size means that Rasenick was cautious about the test’s widespread efficacy. Still, he said, such blood tests can be done on-site at a primary care physician’s office, which could make them easily accessible.

Smoller, the Harvard psychiatric professor, belongs to a group of researchers studying a field called precision psychiatry, which seeks to tailor medicine to the individual rather than rely on a one-size-fits all approach. This might mean, to give one example, identifying which biological characteristics—if any—indicate whether a person will respond to a certain antidepressant. With that information, a psychiatrist could target a patient’s treatment and avoid the yearslong experimentation faced by people like Lowry Smith.

“One of the obstacles we’ve had is this issue of trial-and-error approach to treatment, and the whole appealing idea of precision medicine, which we think of as kind of accounting for or leveraging individual differences in people’s biology, lifestyle and environment, to better diagnose, treat and prevent disease,” Smoller said. “That’s sort of the whole concept.”

To that end, Smoller directs the the Center for Precision Psychiatry at Massachusetts General Hospital, where he oversees research projects such as developing an algorithm to identify those at risk for treatment-resistant depression and better identifying genetic indications of major depressive disorder. But that project will likely take years. 
Still, because of both technological advancements and the expanding range of treatment options, including ketamine, Smoller told The Dispatch he is hopeful about the medical field’s prospects of treating depression: “There is a lot of suffering and a lot of burden that could be prevented if we had better ways of approaching it.”