🚨 PRACTICE-CHANGING STUDY ALERT

Should we use clozapine after just ONE failed antipsychotic?

A new RCT in first-episode psychosis says yes—showing a massive 62.5% response rate in early non-responders.

Early clozapine works, but real-world friction is real 🧵 🤔 The Two-Fail Dogma
For decades, consensus guidelines have dictated that clozapine is strictly for Treatment-Resistant Schizophrenia—defined as failing *two* adequate trials.

But in 2011, Agid et al. showed a grim reality: if a patient fails their first antipsychotic trial, the second trial only yields a ~16% response rate.

Meanwhile, delaying clozapine increases mortality and lowers the chance of functional recovery.

Are we waiting too long?

🚨 ADHD isn't one thing - it's THREE

A massive neuroimaging study just identified 3 neurological "biotypes" of ADHD.

DSM-5 behavioral categories fit into brain-based subtypes - just in time for the new DSM.

Could this explain the huge heterogeneity seen across patients? 🧵 🤔 Background

For decades, we’ve diagnosed ADHD using behavioral checklists. You hit the symptom threshold, you get labeled "inattentive," "hyperactive," or "combined."

But we've known since the 1990s that these consensus-based categories mask enormous biological heterogeneity.

When we average all ADHD brains together in standard imaging studies, the distinct signals wash out.

Recently, the field pivoted to "normative modeling" — essentially pediatric growth charts, but for brain networks.

Instead of just comparing groups, we map how an individual's brain deviates from expected developmental trajectories.

🚨 LANDMARK STUDY ALERT

The definitive 2019 network meta-analysis of antipsychotics just got a massive update—now with the muscarinic modulator KarXT (Cobenfy) and 78,000+ patients.

Clozapine's still best, Cobenfy looks strong, but side effects dictate choice 🧵 🤔 Background

For 70+ years, every approved antipsychotic blocked dopamine D2 receptors. We accepted the metabolic syndrome, weight gain, and EPS as the tragic cost of doing business in treating schizophrenia.

In 2019, the landmark Huhn et al. network meta-analysis (NMA) mapped out the efficacy of 32 oral D2 blockers, but we still lacked a novel mechanism of action.

Fast forward to 2024: the FDA approved xanomeline-trospium (KarXT, commercially known as Cobenfy), a first-in-class M1/M4 muscarinic agonist.

With MANY fast followers coming soon in the pipeline (I'm talking about nearly every company operating in the schizophrenia space with some kind of muscarinic modulator)...

...the field needed a new NMA to see how this entirely new mechanism stacks up against the old guard.

🚨 The DSM is getting a major upgrade.

The APA is officially mapping out how to integrate social determinants of health and intersectionality directly into diagnosis.

In short: context is becoming as important as criteria.

A continuation of our discussion of the new DSM🧵 🤔 Background

For decades, the DSM has essentially been a symptom checklist. We historically stripped away context to boost diagnostic reliability.

In 1980, DSM-III introduced Axis IV for psychosocial stressors, which later morphed into the V/Z-codes in DSM-5.

But despite our biological advances, the overall global burden of mental illness hasn't dropped since 1990.

Patients with serious mental illness still die 15-25 years earlier than the general population.

Biology isn't enough - we have to measure the environment.

🚨 Electroconvulsive therapy (ECT) - the most controversial treatment in Psychiatry

A study of >1 million patients with a depressive disorder re-examined the benefits of ECT.

Here's what you need to know about this landmark paper published this week in @BMJMentalHealth ⚡️ What is ECT?

At its core, ECT is a controlled seizure.

It increases the key growth factor in the brain called BDNF and resets pathologic brain circuits (valleys) toward healthy states.

The procedure itself has become remarkably safe, a far cry from how it has been depicted in the media.

But its longer-term safety, especially on cognition, have been hotly debated (more on this later).

🚨 How "genetic" is ADHD?

We have known for a long time that ADHD is highly heritable
» Twin-based: ~80%
» SNP-based: ~20%

But the exact genes have been elusive.

For the first time, we now have a list of causal genes, discovered from sequencing ~9,000 patients w/ ADHD 🧵 🪙 Rare genetic variants

When we know a disease is heritable but can't find the gene(s) involved, we often look for rare variants.

Why?
» Common: small effect
» Rare: large effect

Although they're rare, when you find one, you can have high confidence they cause the disease.

🚨 FDA'S NEW PLAUSIBLE MECHANISM PATHWAY

@DrMakaryFDA @VPrasadMDMPH just published the @US_FDA most important position piece on biotech regulation in America.

It's short.

but it's 🔥

What you need to know about where the US rare/incurable-disease biotech is headed 🧵 "...a path to market entry for products where a randomized trial is not feasible."

While RCTs are the "gold-standard" for assessing efficacy, it's not always feasible to run or power.

FDA typically requires two "successful" RCTs for approval.

It's now going to be flexible.

🚨 A WEEK OF LANDMARK PAPERS

Clozapine is the batman of psychiatric meds.

A villain in the public eye. A hero behind the scenes.

Study after study, clozapine has been shown to be the most effective antipsychotic for schizophrenia.

But what about for other disorders? 🧵 🗺️ Registry studies

Sweden and Finland have national registries for healthcare.

That means every patient's data is accessible to researchers.

An entire population.

The huge sample size allows us to answer questions in psychiatry that nobody would run a trial for.

🚨 Are SSRIs actually effective in real-world settings?

I'm talking about:
» Outpatient
» Mild depression
» Prescribed by family medicine/GPs

There is one, 500+ patient study that actually looked at this question - the 2019 PANDA study.

The answer?

NO... and... sort of?🧵 📊 Why another SSRI trial?

Network meta-analysis of antidepressants show that they simply work.

But meta-analyses are only as good as the trials that go into them

And usually, those trials are:
» Industry-sponsored
» Sub-specialty settings
» Moderate-severe depression

🚨 LANDMARK STUDY ALERT

The largest, multicenter, assessor-blinded, RCT of antipsychotic efficacy was just published this week.

7 arms.

7 antipsychotics.

3000 patients.

» All antipsychotics ↓ symptoms by 50% in 6 weeks
» Olanzapine, risperidone > the rest BUT

and it's a big BUT

The efficacy difference between the best (Olanzapine) and worst (Ziprasidone) was only an absolute percentage reduction of 8%.

Meanwhile, Olanzapine had an OUTSIZED impact on weight gain and metabolic syndrome.

🚨 Is ketamine effective for treating depression when tested in a trial that actually controls for confounders?

You be the judge:
» 3 point difference on MADRS vs control
» MADRS is a 60 point scale

How to reconcile this landmark study published yesterday in JAMA Psych 🧵 💉What to know about ketamine for depression

Ketamine has time and time again shown to be effective for treating depression:
✅ Treatment-resistance
✅ Bipolar depression
✅ Suicidality ↓

🚨 The older the sperm, the more likely it is to cause autism

Sequencing sperm using nearly error-free sequencing → measure % sperm carrying disease-causing mutations:

2% at age 30 → 4.5% at 70

Most were autism genes

New paper in @Nature today 🧵 👴 Why does parental age matter?

• Spermatogonial stem cells produce millions of sperm daily, with a mutation rate 5–20x lower than other cells.
• Driver mutations in these cells can clonally expand, increasing their presence in sperm and potentially passing to offspring.
• Prior studies linked 13 genes to developmental disorders, but new tech (NanoSeq) now allows broader exploration.

x.com/KeithSakata/st…

🚨 The FIRST RCT of antipsychotic discontinuation was published YESTERDAY.

Do antipsychotics in patients with first-episode psychosis actually improve long-term outcomes?

In short:

NO

But it's nuanced.

Here's what you need to know about this landmark paper in JAMA Psych🧵 🤔 To Continue or Taper?

Antipsychotics (APs) are a staple of preventing relapse in Schizophrenia, but long-term studies are mixed.

Wunderlink et al. (% with functional recovery):
» Continue: 17.6%
» Discontinue or dose reduce (DRD): 40.4%

Chen et al. (% patients with poor outcomes):
» Continue: 21%
» DRD: 39%

Neither study was a randomized controlled trial

🌤️ Cognitive Behavioral Therapy

You hear it all the time.

But what is it actually?

And why is it so effective?

Therapy is shrouded with mystical know-how that doesn't need to be complicated.

Here are the five core concepts you need to know to start CBT right now 🧵 📐 The Cognitive Triangle

The core idea of CBT is the Cognitive Triangle:

An event happens that triggers us → what happens?
• Automatic thoughts ♻️Emotions ♻️Behaviors

CBT is the process of recognizing this cycle → breaking it → leveraging it

🚨 Cannabis was once thought harmless, with boundless benefits.

But since 2005 to today:
» Cannabis use disorder: ↑500%
» Psychosis not otherwise specified: ↑84%
» Schizophrenia: ↓27%..??

Explaining the paradox of this study of 13 MILLION people published this year 🧵 📔 Schizophrenia is diagnosed based on the DSM-5 criteria

» Positive symptoms: psychosis, delusions, disorganization
» Negative symptoms: alogia, avolition, emotional blunting

Crucially, Criterion E states: "not attributable to the physiological effects of a substance"

🚨Should we put Lithium in the water supply?🚨

Microdosing Lithium may protect against Alzheimer's disease.

But not just any Lithium - a very specific form of a Lithium salt.

Here's what you need to know about the @Nature paper the anti-aging community is talking about 👇 💊Lithium was an accident

In the 1940s, an Australian psychiatrist John Cade accidentally discovered anti-manic properties of Lithium.

He had this theory that bipolar disorder could be transmitted via toxins in the urine.

So, he took urine from patients with mania and injected them into guinea pigs.

The guinea pigs died.

He realized that what was killing the mice was the highly concentrated uric acid that was crystallizing into urate. At that time, it was well known that Lithium could help dissolve urate crystals.

So, John Cade tried the experiment again, this time adding Lithium to the patients' urine before injecting into guinea pigs.

The result? The guinea pigs got sleepy.

John Cade interpreted these results to mean that Lithium might have calming or anti-manic properties.

After first testing Lithium on himself, he then trialed it in 10 manic patients, leading to the landmark discovery of Lithium as an anti-manic medication.

Definitely some cowboy psychiatry...

How do the hundreds of genetic risk-factors for #Schizophrenia actually lead to disease? New work medrxiv.org/content/10.110… from @manoliskellis illuminates the #SingleCell landscape of #Brains from patients with #Schizophrenia and their healthy counterparts🔬🧠🧵1/10 Combining single-nuclear RNA-Seq @10xGenomics, barcode #ing, and multi-level cell-state decomposition, first author Brad Ruzicka et al. identify 20 cell types in the prefrontal cortex of 24 patients with #Schizophrenia and 24 controls 2/10

Why is it becoming harder to #remember things you have recently read? New work nature.com/articles/s4158… out in @nature from Anthony Wagner's group @Stanford explore how multi-media #multitasking impacts a specific axis of #attention that in turn, impacts our #memory📱🧠🧵1/11 When we're reading a book and simultaneously checking our #twitter or #finsta feeds, we're not actually multitasking - we're task-switching! We switch our focus from one goal to another, interrupting our train of thought and scattering our #attention 2/11