How the drug was found

looking for: a drug to blunt surgical shock · accident

A French navy surgeon, Henri Laborit, was testing an antihistamine derivative made by Rhone-Poulenc chemists in December 1950 to protect patients from surgical shock. He saw that patients became calm and indifferent without losing consciousness, and pushed psychiatrists to try it. In 1952 two Paris groups reported it calmed agitated psychotic patients.

note: Which psychiatric group deserves the credit is contested in the literature, so this is a story about a compound, not one hero.

• Lopez-Munoz F, et al. History of the discovery and clinical introduction of chlorpromazine. Ann Clin Psychiatry. 2005. PMID 16433053.
• Ban TA. Fifty years chlorpromazine: a historical perspective. Neuropsychiatr Dis Treat. 2007. PMID 19300578.

looking for: a toxin in the urine of manic patients · accident

John Cade, working in a repatriation hospital near Melbourne, believed mania was poisoning by something the body excreted. He injected guinea pigs with urine from manic patients, and to test uric acid he needed a soluble form, so he used lithium urate. Instead of making the animals worse, it made them placid. He gave lithium to ten manic patients and reported in 1949 that their excitement came under control.

note: The theory was wrong and the result was right, which one historian calls a false rationale leading to a correct empirical finding. Lithium is an element, in the same chemical family as the sodium in table salt, which is why nobody owns it and nobody advertises it.

• Cade JF. Lithium salts in the treatment of psychotic excitement. Med J Aust. 1949. PMID 18142718.
• Shorter E. The history of lithium therapy. Bipolar Disord. 2009. PMID 19538681.
• Ban TA. The role of serendipity in drug discovery. Dialogues Clin Neurosci. 2006. PMID 17117615.

looking for: a tuberculosis drug · accident

Made at Roche in 1951 as a tuberculosis treatment. At Sea View Hospital on Staten Island, doctors treating tuberculosis patients noticed in 1952 that the patients became euphoric and overactive. By 1957 psychiatrists were testing it in depression and calling it a psychic energizer.

note: The often-repeated image of patients dancing in the hospital halls comes from press coverage; the journal record says euphoric and overactive.

• Loomer HP, Saunders JC, Kline NS. A clinical and pharmacodynamic evaluation of iproniazid as a psychic energizer. Psychiatr Res Rep. 1957. PMID 13542681.
• Lopez-Munoz F, et al. Half a century of antidepressant drugs, part I: monoamine oxidase inhibitors. J Clin Psychopharmacol. 2007. PMID 18004120.

looking for: a new antipsychotic · accident

The Swiss company Geigy hoped compound G 22355 would work like chlorpromazine and sent it to Roland Kuhn at a remote asylum. It failed in schizophrenia. Before sending the supply back, Kuhn tried it in a woman with severe depression and in January 1956 saw her improve. He confirmed it the hard way: patients relapsed when the drug was stopped and recovered when it was restarted.

note: Kuhn's method was open-ended clinical observation, not a controlled trial. The finding was later confirmed in trials.

• Kuhn R. The treatment of depressive states with G 22355 (imipramine hydrochloride). Am J Psychiatry. 1958. PMID 13583250.
• Brown WA, Rosdolsky M. The clinical discovery of imipramine. Am J Psychiatry. 2015. PMID 25930134.

looking for: a tranquilizer, from old dye chemistry · accident

At Roche in New Jersey, Leo Sternbach revisited dye compounds he had studied as a student in the 1930s, hoping one would be a tranquilizer. The series was inert and one compound was shelved untested. During a lab cleanup in 1957 it was found and sent for screening, and it calmed animals and relaxed their muscles. Only then did the chemists work out it was a new kind of ring, a benzodiazepine. It was sold as Librium in 1960.

note: One historian's summary of the discovery is two words: sheer luck. The benzodiazepines that followed, including diazepam, alprazolam, clonazepam, lorazepam and temazepam, were designed on purpose from this one.

• Lopez-Munoz F, et al. The discovery of chlordiazepoxide and the clinical introduction of benzodiazepines. J Anxiety Disord. 2011. PMID 21315551.
• Wick JY. The history of benzodiazepines. Consult Pharm. 2013. PMID 24007886.
• Sternbach LH. The benzodiazepine story. J Med Chem. 1979. PMID 34039.

looking for: relief for headaches after a spinal procedure · accident

Amphetamine was made in the late 1920s as a substitute for ephedrine, an asthma drug. In 1937 the psychiatrist Charles Bradley gave Benzedrine to children in a Rhode Island residential home to ease the headaches that followed a diagnostic spinal procedure. The headaches did not improve. About half the children showed strikingly better schoolwork and calmer behavior. The finding was largely ignored for about twenty-five years.

note: This is the root of stimulant treatment for what is now called ADHD, found by a doctor who was treating something else.

• Strohl MP. Bradley's Benzedrine studies on children with behavioral disorders. Yale J Biol Med. 2011. PMID 21451781.
• Lange KW, et al. The history of attention deficit hyperactivity disorder. Atten Defic Hyperact Disord. 2010. PMID 21258430.
• Rasmussen N. Making the first anti-depressant: amphetamine in American medicine, 1929 to 1950. J Hist Med Allied Sci. 2006. PMID 16492800.

looking for: a new antipsychotic · accident

Chemists at Mead Johnson reported the series in 1969 under the heading psychosedative agents, a search for antipsychotics. Buspirone did not work as an antipsychotic. It reduced anxiety without sedation and without touching the receptor that benzodiazepines act on, which opened a new class.

note: A designed molecule that found a different job.

• Taylor DP. Buspirone, a new approach to the treatment of anxiety. FASEB J. 1988. PMID 2836252.
• Eison AS, Temple DL. Buspirone: review of its pharmacology and current perspectives on its mechanism of action. Am J Med. 1986. PMID 2870639.

looking for: a shorter-acting anesthetic; later, a model of schizophrenia · accident

Made in 1962 as a shorter-acting relative of phencyclidine, which had failed as an anesthetic. Edward Domino tested it in twenty volunteers from a prison population in 1964 and 1965 and coined the term dissociative anesthetic. The depression finding came thirty-five years later from a Yale lab that had been giving small doses to volunteers to mimic schizophrenia. In 2000 a seven-patient trial saw depression scores fall within seventy-two hours of a single infusion.

note: Ketamine was never a horse tranquilizer first; that is a legend. Esketamine, approved in 2019, is one half of the ketamine molecule.

• Domino EF. Taming the ketamine tiger. 1965. Anesthesiology. 2010. PMID 20693870.
• Berman RM, et al. Antidepressant effects of ketamine in depressed patients. Biol Psychiatry. 2000. PMID 10686270.

looking for: a new antipsychotic in the chlorpromazine mold · accident, then a scare

Synthesized in 1958 at a Swiss company as part of a tricyclic series, clozapine was pharmacologically strange: it treated psychosis but did not cause the movement side effects that were then taken as proof a drug was an antipsychotic, so some pharmacologists refused to believe it was one. In 1975 a cluster of deaths from a blood disorder in Finland led to its withdrawal in most of the world. Psychiatrists kept demanding it for patients nothing else helped, and it returned in the United States in 1990 under mandatory blood monitoring, ending a fifteen-year gap in new antipsychotics.

note: The line that the company was looking for an antidepressant is not in any source we read. Olanzapine and quetiapine were designed afterward to keep its benefits without the blood risk.

• Hippius H. A historical perspective of clozapine. J Clin Psychiatry. 1999. PMID 10372606.
• Crilly J. The history of clozapine and its emergence in the US market: a review and analysis. Hist Psychiatry. 2007. PMID 17580753.
• Shen WW. A history of antipsychotic drug development. Compr Psychiatry. 1999. PMID 10579370.

looking for: an epilepsy drug that lowers folate · designed for a wrong mechanism; mood use an accident

In 1973 a Lancet paper proposed that anticonvulsants work partly by lowering folate, so chemists at Wellcome screened folate-blocking compounds for anticonvulsant effect. Lamotrigine was a good anticonvulsant and a weak folate blocker: it worked, but not for the reason it was built. Its use in bipolar disorder came from epilepsy doctors noticing their patients' moods improved, one psychiatrist trying it in two patients nothing else had helped, and a company willing to gamble on a field with no new drugs in decades.

note: The authors of its own history called it a story of serendipity, clinical observation and risk taking.

• Reynolds EH. Anticonvulsants, folic acid, and epilepsy. Lancet. 1973. PMID 4126153.
• Weisler RH, Calabrese JR, et al. Discovery and development of lamotrigine for bipolar disorder: a story of serendipity, clinical observations, risk taking, and persistence. J Affect Disord. 2008. PMID 18001843.

looking for: a way to get GABA into the brain · designed for a wrong mechanism

GABA, the brain's main calming transmitter, cannot cross from the blood into the brain. German chemists built it onto a ring to smuggle it in, and the chemist's own 1994 account is titled antiepileptics from GABA. It worked as an anticonvulsant. Then it turned out gabapentin does not act on GABA receptors at all; it binds part of a calcium channel nobody was aiming for, which is also why it helps nerve pain.

note: If someone tells you gabapentin works on GABA, that is the original theory, not the finding.

• Satzinger G. Antiepileptics from gamma-aminobutyric acid. Arzneimittelforschung. 1994. PMID 8192689.
• Taylor CP, et al. A summary of mechanistic hypotheses of gabapentin pharmacology. Epilepsy Res. 1998. PMID 9551785.

looking for: nothing in particular; a mouse screen hit · accident

In 1974 a small French company found a compound that made mice more active without the usual stimulant effects on the body. A Lyon sleep scientist, Michel Jouvet, gave it to narcolepsy patients with mixed results, and blood tests then revealed that the body converted it into a second compound that was doing the work: modafinil. In 1983 Jouvet's group gave modafinil itself to patients and saw sleep attacks fall sharply. The company was at first not interested in developing it; the scientist's insistence pushed it into trials, and France approved it in 1992.

note: The marketed drug is a metabolite the company had not planned on.

• Billiard M, Broughton R. Modafinil: its discovery, the early European and North American experience in the treatment of narcolepsy and idiopathic hypersomnia, and its subsequent use in other medical conditions. Sleep Med. 2018. PMID 30174215.
• Bastuji H, Jouvet M. Successful treatment of idiopathic hypersomnia and narcolepsy with modafinil. Prog Neuropsychopharmacol Biol Psychiatry. 1988. PMID 2906157.

looking for: a blood-pressure drug that would help memory without sedation · repurposed from a monkey experiment

Guanfacine was a blood-pressure pill. At Yale in 1988, researchers testing aged rhesus monkeys found that low doses improved the monkeys' working memory without the sedation or blood-pressure drop that a related drug caused. In 1995 the same group ran an open trial in thirteen children with ADHD, and a placebo-controlled trial followed in 2001.

• Arnsten AF, Cai JX, Goldman-Rakic PS. The alpha-2 adrenergic agonist guanfacine improves memory in aged monkeys without sedative or hypotensive side effects. J Neurosci. 1988. PMID 2903226.
• Hunt RD, Arnsten AF, Asbell MD. An open trial of guanfacine in the treatment of attention-deficit hyperactivity disorder. J Am Acad Child Adolesc Psychiatry. 1995. PMID 7860456.

looking for: an antidepressant · a shelved antidepressant, repurposed

Made at Lilly by the same group that produced fluoxetine and published in 1982 as a clean norepinephrine uptake blocker, tomoxetine was tried as an antidepressant in a small 1984 study of ten inpatients. Development for depression did not continue; the published record does not say why. Fourteen years later a Boston group tested it in twenty-two adults with ADHD in a crossover trial, eleven improved on the drug against two on placebo, and it came back as atomoxetine.

note: The story that it was renamed to avoid confusion with the cancer drug tamoxifen has no journal source.

• Chouinard G, et al. An early phase II clinical trial of tomoxetine (LY139603) in the treatment of newly admitted depressed patients. Psychopharmacology. 1984. PMID 6429697.
• Spencer T, Biederman J, et al. Effectiveness and tolerability of tomoxetine in adults with attention deficit hyperactivity disorder. Am J Psychiatry. 1998. PMID 9585725.

looking for: a new stimulant · designed

Synthesized at Ciba in Switzerland in 1944 by Leandro Panizzon and sold as Ritalin from 1954, named after his wife Marguerite, known as Rita. Its first uses were fatigue, lethargy and narcolepsy; a 1963 controlled trial in children began its ADHD career.

note: The story that Rita took it before tennis and played better is repeated everywhere and traced to no journal source. We do not tell it.

• Lange KW, et al. The history of attention deficit hyperactivity disorder. Atten Defic Hyperact Disord. 2010. PMID 21258430.
• Conners CK, Eisenberg L. The effects of methylphenidate on symptomatology and learning in disturbed children. Am J Psychiatry. 1963. PMID 14051237.

looking for: an antidepressant unlike the tricyclics · designed, then two unplanned turns

Approved in the United States in 1985, withdrawn in 1986 after seizures appeared at high doses, and brought back in 1989 with a dose cap. Its mechanism was still described as unclear a decade later. Clinicians noticed patients on it lost interest in cigarettes, and a 1997 Mayo Clinic trial in smokers who were not depressed found quit rates rose with the dose while depression scores did not change.

note: The named-doctor eureka story about the smoking effect rests on conference abstracts and retellings, so we say clinicians noticed.

• Davidson J. Seizures and bupropion: a review. J Clin Psychiatry. 1989. PMID 2500425.
• Hurt RD, et al. A comparison of sustained-release bupropion and placebo for smoking cessation. N Engl J Med. 1997. PMID 9337378.

looking for: a non-tricyclic antidepressant · designed, with a surprise

Pfizer chemists were making a family of compounds descended from an earlier one dropped for stimulant-like effects. They found the two mirror-image shapes of the molecule did opposite things: one shape blocked serotonin uptake, the other blocked dopamine and norepinephrine. The serotonin-selective shape became sertraline, published in 1983.

note: Designed on purpose; the surprise was which shape did what.

• Welch WM, et al. Nontricyclic antidepressant agents derived from cis- and trans-1-amino-4-aryltetralins. J Med Chem. 1984. PMID 6492080.
• Koe BK, et al. Sertraline, a new uptake inhibitor with selectivity for serotonin. J Pharmacol Exp Ther. 1983. PMID 6310078.

looking for: a drug that blocks serotonin uptake and nothing else · designed

Lilly's Bryan Molloy and David Wong set out in the early 1970s to find exactly that, starting from an antihistamine scaffold. Compound 110140 was serotonin-selective in the 1974 paper. Approval came thirteen years later, in 1987.

note: One of the first psychiatric drugs designed for a chosen target. The stories that it was first meant for weight or blood pressure are not how its discoverers describe it.

• Wong DT, Perry KW, Bymaster FP. The discovery of fluoxetine hydrochloride (Prozac). Nat Rev Drug Discov. 2005. PMID 16121130.

looking for: a serotonin-selective antidepressant · designed

Lundbeck chemists working around a norepinephrine drug of their own found a close relative that did the opposite: a potent, selective serotonin uptake inhibitor, published in 1977. Escitalopram is one of the two mirror-image halves of citalopram, isolated in the 1990s.

note: The story that the citalopram ring came from a failed reaction appears in books and interviews, not in a journal we could find. The claim that the other half of the molecule works against the active half is the manufacturer's.

• Christensen AV, et al. Pharmacology of a new phthalane (Lu 10-171), with specific 5-HT uptake inhibiting properties. Eur J Pharmacol. 1977. PMID 12988.
• Sanchez C, et al. Escitalopram versus citalopram: the surprising role of the R-enantiomer. Psychopharmacology. 2004. PMID 15160261.

looking for: a serotonin-selective antidepressant · designed

A small Danish company, Ferrosan, ran its own serotonin program in parallel with Lilly and Lundbeck; its first candidate was published in 1975 and paroxetine followed from the same program.

note: A program, not a person; no journal tells a personal anecdote.

• Lopez-Munoz F, Alamo C. Monoaminergic neurotransmission: the history of the discovery of antidepressants. Curr Pharm Des. 2009. PMID 19442174.

looking for: a non-tricyclic that blocks two transmitters · designed

Wyeth chemists built and screened a series for norepinephrine and serotonin uptake, deliberately dropping the side-effect targets of the tricyclics; the compound reported in 1986 became venlafaxine. Desvenlafaxine is its main metabolite, developed as its own drug in 2006 as the patent aged.

note: The claim that venlafaxine came out of an opioid painkiller program is repeated but not stated in any journal we found.

• Muth EA, et al. Antidepressant biochemical profile of the novel bicyclic compound Wy-45,030. Biochem Pharmacol. 1986. PMID 3790168.
• Deecher DC, et al. Desvenlafaxine succinate: a new serotonin and norepinephrine reuptake inhibitor. J Pharmacol Exp Ther. 2006. PMID 16675639.

looking for: a dual-action successor to fluoxetine · designed

The same Lilly group that made fluoxetine modified its scaffold to block both serotonin and norepinephrine uptake, reasoning that two actions might beat one; the compound was reported in 1988.

• Wong DT. Duloxetine (LY 248686): an inhibitor of serotonin and noradrenaline uptake and an antidepressant drug candidate. Expert Opin Investig Drugs. 1998. PMID 15991911.

looking for: a better version of the company's own antidepressant · designed

Organon altered its drug mianserin by one atom; the new compound lost most of the uptake-blocking effect and kept a different action, which became its whole identity.

• de Boer TH, et al. Neurochemical and autonomic pharmacological profiles of the 6-aza-analogue of mianserin, Org 3770 and its enantiomers. Neuropharmacology. 1988. PMID 3419539.

looking for: one molecule with several serotonin actions · designed

Lundbeck set out to combine transporter blockade with direct receptor actions in a single compound and published the result in 2011. Pure medicinal chemistry.

• Bang-Andersen B, et al. Discovery of Lu AA21004: a novel multimodal compound for the treatment of major depressive disorder. J Med Chem. 2011. PMID 21486038.

looking for: a drug for the mental pain of depression · designed from a theory

Bruno Silvestrini's group in Rome built trazodone on the idea that depression involves a heightened perception of unpleasant experience, and chose it using a pain screen in animals rather than the usual antidepressant tests. Today it is prescribed for sleep far more than for depression.

note: A rare case of a drug designed from a theory of the illness.

• Silvestrini B, Valeri P. Trazodone, a new avenue in the treatment of depression. Psychopathology. 1984. PMID 6144138.

looking for: a sleep drug that was not a benzodiazepine · designed

Chemists at Synthelabo in Paris deliberately built a new chemical family for sleep, and the 1985 report showed it induced slow-wave sleep in rats without the benzodiazepine pattern.

• Arbilla S, et al. Pharmacological profile of the imidazopyridine zolpidem at benzodiazepine receptors and electrocorticogram in rats. Naunyn Schmiedebergs Arch Pharmacol. 1985. PMID 2997631.

looking for: variations on the accidental first one · designed

Every one was built on purpose from the Librium ring: diazepam as a more potent analog (sold 1963), lorazepam for cleaner metabolism, clonazepam as an anticonvulsant first, temazepam as a short-acting sleep drug, alprazolam with an extra ring fused on. No accidents, and no anecdotes worth telling.

• Wick JY. The history of benzodiazepines. Consult Pharm. 2013. PMID 24007886.
• Hester JB, Rudzik AD, VonVoigtlander PF. 6-Phenyl-4H-s-triazolo[4,3-a][1,4]benzodiazepines which have central nervous system depressant activity. J Med Chem. 1980. PMID 6103958.

looking for: an antihistamine · designed

A Belgian antihistamine introduced in 1956 and marketed from the start as a calming agent. That is all the journal record supports.

note: No journal tells who made it or what they were looking for, so we tell no story.

• Hutcheon DE, et al. The pharmacology of hydroxyzine. J Pharmacol Exp Ther. 1956. PMID 13385806.

looking for: clozapine's benefits without its blood risk · designed

Every one was built on purpose. Paul Janssen's lab had noticed that adding a serotonin blocker to a dopamine blocker changed the side-effect picture, and risperidone put both in one molecule, reported in 1988. Olanzapine was built at Lilly to copy clozapine's skeleton with one ring swapped. Quetiapine was chosen by testing clozapine relatives in monkeys for which ones did not cause dyskinesia. Ziprasidone was engineered for a target ratio of serotonin to dopamine blockade. Aripiprazole was designed at Otsuka to act on the dopamine receptor only partway. Lurasidone has no discovery narrative in the journals at all. From the late 1980s on, antipsychotics came from design, not from noticing.

note: The story that Lilly's first candidate was dropped for liver toxicity before olanzapine appears in no journal we found. Quetiapine was not a lucky find; the monkey test was the method.

• Janssen PA, et al. Pharmacology of risperidone (R 64 766), a new antipsychotic with serotonin-S2 and dopamine-D2 antagonistic properties. J Pharmacol Exp Ther. 1988. PMID 2450200.
• Warawa EJ, et al. Behavioral approach to nondyskinetic dopamine antagonists: identification of seroquel. J Med Chem. 2001. PMID 11462978.
• Kikuchi T, et al. Discovery research and development history of the dopamine D2 receptor partial agonists, aripiprazole and brexpiprazole. Neuropsychopharmacol Rep. 2021. PMID 33960741.

looking for: a strong painkiller that does not addict · designed as a painkiller; addiction use a repurposing

A British company spent the 1960s chasing a goal a century old: morphine-strength pain relief without addiction. Buprenorphine came out of that work with an unusual property, a ceiling on its own effect. In 1978 a researcher at the federal addiction research center in Lexington, Kentucky, reported that in former addicts it blocked morphine's effects for nearly thirty hours with little physical dependence, and proposed it as a treatment for addiction. The United States approved that use twenty-four years later.

• Jasinski DR, Pevnick JS, Griffith JD. Human pharmacology and abuse potential of the analgesic buprenorphine: a potential agent for treating narcotic addiction. Arch Gen Psychiatry. 1978. PMID 215096.
• Campbell ND, Lovell AM. The history of the development of buprenorphine as an addiction therapeutic. Ann N Y Acad Sci. 2012. PMID 22256949.

looking for: an amphetamine that is harder to misuse · designed

Amphetamine with the amino acid lysine bonded to it, so it is inactive until enzymes in red blood cells cut the two apart. That slows and caps the rise of amphetamine in the blood and makes snorting or injecting it pointless; a 2009 study in adult stimulant users found intravenous lisdexamfetamine scored no higher than placebo on liking.

note: The company's founding story and the inventors' motives are not in the journal literature, so we tell only the mechanism.

• Jasinski DR, Krishnan S. Human pharmacology of intravenous lisdexamfetamine dimesylate: abuse liability in adult stimulant abusers. J Psychopharmacol. 2009. PMID 18635707.