Chapter 1
Psychopharmacological history through the looking-glass of advancing years
David Healy
Introduction
In 1956, Roland Kuhn described the effects of imipramine in a 56-year-old woman, Paula J F, who had a severe delusional melancholic disorder. Imipramine was subsequently demonstrated to be effective in a patient group suffering from vital depression with psychomotor retardation and often psychotic developments in older individuals. At the time, this condition was regarded by many as the distinct diagnostic entity of involutional melancholia. The success of imipramine led to an almost immediate disappearance of this syndrome.
It might not be thought surprising that an effective psychotropic agent should lead to the disappearance of a syndrome, on the basis that most people assume this is what effective agents should do. What was astonishing about this disappearance was that, far from imipramine being effective in the sense that penicillin was for general paralysis of the insane (GPI), imipramine seemed to be a treatment that validated the existence of the syndrome and the need for its possible ongoing maintenance treatment. Yet a few years later the diagnostic category of involutional melancholia had effectively disappeared.
This disappearance is at odds with the mainstream of psychopharmacological history, in which classification has been driven largely by the response (or otherwise) to psychotropic drugs rather than by the clinical criteria that drove classification in the pre-psychotropic era. The disappearance of involutional melancholia gives a first hint that the history of psychotropic agents given to older individuals might be radically different to conventional histories of the psychotropic era. Unfortunately, there are few histories of what happens when psychotropic drugs are given to older individuals.1,2 The usual assumption appears to have been that what happens in the elderly will only reflect what is happening in a general adult group. However, there are good grounds to think that, in future, the books may reflect a quite different history, as this brief chapter seeks to illustrate.
The conventional history of the antidepressants
The conventional origin of the antidepressant was when Roland Kuhn discovered the effects of imipramine in 1956. Unpersuaded of his reports, Geigy sought confirmation elsewhere. At a time when compounds could be synthesised in the laboratory and could enter mainstream clinical use within 3 months, imipramine took 2 years to market. A major reason for Geigyâs reluctance appears to have stemmed from the perception that depression was a relatively rare condition.3
At almost the same time as Kuhn reported on the effects of imipramine to Geigy, Nathan Kline made the discovery that one of two recently synthesised anti-tubercular agents, iproniazid, had psychic energising properties. It later became known as one of the first of the monoamine oxidase inhibitor (MAOI) antidepressants. Roche were not interested in such a compound.
Yet even before Kline or Kuhn, in 1952, Max Lurie had discovered that isoniazid, another anti-tubercular agent, had antidepressant properties. This discovery was replicated in Paris in the same year. However, none of the pharmaceutical companies that Lurie approached were interested in the idea of an antidepressant. The very term was new, having probably been coined by Lurie.
There had indeed been yet another discovery of a drug with antidepressant properties before Kuhn and Kline. In 1955, researchers from the Institute of Psychiatry demonstrated in an article published in the Lancet that reserpine was an antidepressant. However, Ciba, the makers of reserpine, were not interested in developing an antidepressant.
It was not until the discovery of amitriptyline and its marketing by Merck that any pharmaceutical company showed an interest in developing an antidepressant. Merck realised that if they were going to market this new class they would also have to sell the illness, so they purchased 50 000 copies of Frank Aydâs Recognising the Depressed Patient.4 Despite this, sales of the antidepressants remained flat during the 1960s and 1970s.
The antidepressant market only truly developed when the benzodiazepine tranquillisers ran into difficulties in the 1980s. Concerns about dependence on these drugs steered the marketing development of a new generation of drugs that acted on the serotonin system towards an antidepressant route.
Thus the selective serotonin reuptake inhibitors (SSRIs) were developed in 1971 from a conjunction of clinical observations by Paul Kielholz and neuroscientific observations by Arvid Carlsson. Kielholz noted that not all of the tricyclic and MAOI antidepressants did the same things to people who were depressed, and he categorised âantidepressantsâ according to whether they were energy enhancing or whether they produced âcognitiveâ changes. Reviewing Kielholzâs classification in the late 1960s, Carlsson, one of the key pioneers of neuroscience, noted that the antidepressants which Kielholz stated were more energy enhancing acted on the catecholamine system, whereas those with cognitive effects acted on the serotonergic system. This led Carlsson to create the first SSRI, namely zimelidine, to see what the effects of such a drug would be. Zimelidine was launched in 1982, but was later withdrawn because of toxicological problems. It was succeeded by fluvoxamine, fluoxetine, sertraline, paroxetine and citalopram.
The unconventional history of the antidepressants
Following the emergence of the SSRIs, the antidepressant market mushroomed. If antidepressants are effective, this should not have happened, and indeed this is not what happened in the case of involutional melancholia. The most parsimonious explanation of what happened was that a process was initiated whereby cases of Valium were converted into cases of Prozac. This did not happen in Japan and many other parts of the world where the SSRIs did not emerge as antidepressants so early on. In these countries the antidepressant market remains a comparatively small one compared with the market for tranquillisers.
The development of the SSRIs was predicated on the notion that agents which act on catecholamine systems were energy enhancing whereas agents which act on the serotonin system appeared to do something else that was either anxiolytic or produced some other cognitive effect. The SSRIs were therefore designed as drugs that were different from the tricyclic antidepressants, rather than âcleaned-upâ versions of the same (which is how they had originally been marketed).
In fact the SSRIs have proved singularly ineffective for hospital depressions or depressions of the kind that Paula J F suffered from. In contrast, selective catecholamine reuptake inhibitors such as reboxetine have proved very effective in hospital depressions, but apparently much less effective in community depressions of the more anxious depressive type. This suggests that there may well be a distinctive component to the kinds of depression cases who end up in hospital, particularly those who end up in hospital in their middle to later years.
There remains a consensus of opinion that electroconvulsive therapy (ECT) is an effective treatment for the kinds of vital depression for which imipramine was first thought to be useful. ECT is not regarded as a treatment for those kinds of anxious depressions which are found in the community. Broadly speaking, the conditions thought to be responsive to ECT are the ones that in the 1980s were thought to be characterised by dexamethasone non-suppression, a set of depressions in which there is thought to be substantial elevation of cortisol levels, possibly brought about by escape from circadian control. This is not found in the depression cases in the community, and SSRIs in general appear to be less effective in dexamethasone non-suppression.5
Recently, however, mifepristone, which was previously used as an abortifacient agent, has been demonstrated to have an effect on severe or psychotic depressions.6 It also appears to have effects on cortisol systems that are comparable to the effects of agents such as dexamethasone or ECT. Should the clinical efficacy of mifepristone be replicated, it will be an interesting test of pharmaceutical company perceptions of the depressive market to see whether it, unlike reboxetine, is developed further. If this were to happen, there might be a need to reinvent melancholia (if not involutional melancholia) and redesignate the SSRIs as anxiolytics, a process that is arguably well under way.
There are alternative frameworks within which to place such a development. While involutional melancholia may never have been the distinctive entity it was once thought to be, this clinical classification arguably embodied some clinical insights of substance. The distinctive response of this syndrome to some agents and not to others points to different physiological underpinnings. One possibility is that a constitutional or temperamental component shapes the differential response to selective agents seen in these vital depressions. Alternatively, a constitutional or temperamental component may shape clinical presentations so that some affective disorders in older years have features of anxiety, while others have features of a vital syndrome. Older age may in fact be a key setting in which some of these issues that seem irresolvable in younger populations may be teased out.
The conventional history of the antipsychotics
The conventional history of the antipsychotics has it that the discovery of chlorpromazine was first interpreted as the discovery of a new form of sleep treatment. The subsequent synthesis of a number of non-sedative phenothiazine agents, such as proclorperazine, led to a recognition that these new agents might have neuroleptic rather than just sedative effects.7
This created two schools of thought. One school based in Lyon argued for a classification of neuroleptic drugs ranging from the sedative to the incisive, and for a recognition that different agents would suit different clinical syndromes and different constitutional types. This notion underpinned the development of sedative neuroleptics such as thioridazine and levomepromazine, in addition to perphenazine and other more incisive agents. In Paris, in contrast, Deniker and Delay argued that a neuroleptic effect was the common therapeutic effect of all of these agents, and that other properties such as sedative properties were simply side-effects. The notions of Delay and Deniker led to a research and drug development agenda that focused on optimising the core neuroleptic effect.
The battle between Lyon and Paris was brief but bitter, with Paris triumphing. The development in 1958 of haloperidol, a non-sedative agent, appeared to endorse the insights of the Paris school. Haloperidol became the ultimate neuroleptic, and subsequent antipsychotics were developed on the basis of a set of laboratory tests aimed at replicating its effects.
The development of haloperidol almost led to the abortion of clozapine. On a number of the emerging tests of neuroleptics the latter drug, which had first been synthesised in 1958, failed. This failure led to a relatively half-hearted development programme, which was abandoned once the propensity of clozapine to produce agranulocytosis was recognised. However, a number of European clinicians, including Jules Angst, Raymond Battegay and Hanns Hippius, objected and argued that clozapine was indeed an effective antipsychotic even though it appeared not to have standard neuroleptic properties. The protest kept clozapine alive.
With the discovery of the effect of the neuroleptic drugs on dopamine systems, the development pathway focused on producing more selective and more potent D2-receptor antagonists. This culminated in the benzamide group of drugs, namely sulpiride, amisulpiride and remoxipride, which arguably had somewhat fewer side-effects than other agents but were in general no more potent. With the re-emergence of clozapine, arguments were offered that the additional 5HT2 blockade which this compound offered might account for its apparently better effects than standard neuroleptics in schizophrenia.
The unconventional history of the antipsychotics
In the mid-1960s, concerns emerged about the propensity of neuroleptics to produce tardive dyskinesia. This syndrome, which is sometimes present in untreated older patients with psychoses, appeared to occur with a much greater frequency and severity in patients who were taking neuroleptics. At a time of rising antipsychiatric discontent with orthodox psychiatry, tardive dyskinesia provided a lightning rod to focus this discontent. In 1974, SmithKline and French settled their first million-dollar legal action with regard to tardive dyskinesia, and a subsequent generation of antipsychotic drug development was abandoned.
Apart from the benzamides, which were not developed in the USA, drug development in this area was only renewed in the wake of the re-emergence of clozapine. Contrary to popular mythologies, the re-emergence of clozapine stemmed not from any superior clinical efficacy, but rather from the fact that this was a compound that did not cause tardive dyskinesia. The agents that came in the wake of clozapine, namely sertindole, olanzapine, quetiapine and risperidone, were also developed primarily as agents that would minimise the risk of tardive dyskinesia.
Apart from a lower incidence of tardive dyskinesia, especially for quetiapine, none of these newer agents are more effective than the older agents. Risperidone and olanzapine both produce dose-dependent extrapyramidal side-effects, and are in fact rather typical neuroleptics. The non-extrapyramidal spectrum of side-effects of these agents in general suggests that they may in many ways be less safe than the older agents. Olanzapine and clozapine are associated with elevated lipid levels, the production of diabetes and weight gain, all of which can be expected to elevate cardiovascular mortality.
The new agents differ in their propensity to block D2 receptors and also in terms of their sedative properties. Amisulpiride and risperidone are most like conventional neuroleptics, while clozapine and quetiapine are more like sedative neuroleptics such as thioridazine or levomopromazine. Recent studies which tested the ability of agents such as amisulpiride to augment the effects of clozapine in clozap...