Atypical Antipsychotics: Mechanism of Action

Transcription

1 In Review Atypical Antipsychotics: Mechanism of Action Philip Seeman, MD, PhD 1 Background: Al though the prin ci pal brain tar get that all an tipsy chotic drugs at tach to is the do pa - mine D 2 re cep tor, tra di tional or typi cal an tipsy chot ics, by at tach ing to it, in duce ex tra py r a mi dal signs and symp toms (EPS). They also, by bind ing to the D2 re cep tor, ele vate se rum pro lac tin. Atypi cal an - tipsy chot ics given in dos ages within the clini cally ef fec tive range do not bring about these ad verse clini cal ef fects. To un der stand how these drugs work, it is im por tant to ex am ine the atypi cal an tipsy - chot ics mecha nism of ac tion and how it dif fers from that of the more typi cal drugs. Method: This re view analy zes the af fini ties, the oc cu pan cies, and the dis so cia tion time- course of vari ous an tipsy chot ics at do pa mine D 2 re cep tors and at se ro tonin (5-HT) re cep tors, both in the test tube and in live pa tients. Results: Of the 31 an tipsy chot ics ex am ined, the older tra di tional an tipsy chot ics such as tri flu pera - zine, pi mozide, chlor pro maz ine, fluphe nazine, ha loperi dol, and flupen thixol bind more tightly than do pa mine it self to the do pa mine D 2 re cep tor, with dis so cia tion con stants that are lower than that for do pa mine. The newer, atypi cal an tipsy chot ics such as queti apine, re moxi pride, clo zap ine, ol an - zap ine, sert in dole, zi pra si done, and amisul pride all bind more loosely than do pa mine to the do pa - mine D 2 re cep tor and have dis so cia tion con stants higher than that for do pa mine. These tight and loose bind ing data agree with the rates of an tipsy chotic dis so cia tion from the human- cloned D 2 re - cep tor. For in stance, ra dio ac tive ha loperi dol, chlor pro maz ine, and ra clo pride all dis s o ci ate very slowly over a 30- minute time span, while ra dio ac tive queti apine, clo zap ine, re moxi pride, and amisul pride dis so ci ate rap idly, in less than 60 sec onds. These data also match clini cal brainimaging find ings that show ha loperi dol re main ing con stantly bound to D2 in hu mans un der go ing 2 posi tron emis sion to mo gra phy (PET) scans 24 hours apart. Con versely, the oc cu pa tion of D 2 by clo zap ine or queti apine has mostly dis ap peared af ter 24 hours. Conclusion: Atypi cals clini cally help pa tients by tran siently oc cu py ing D 2 re cep tors and then rap idly dis so ci at ing to al low nor mal do pa mine neu ro trans mis sion. This keeps pro lac tin lev els nor mal, spares cog ni tion, and ob vi ates EPS. One the ory of atypi cal ity is that the newer drugs block 5-HT 2A re cep - tors at the same time as they block do pa mine re cep tors and that, some how, this serotonin- dopamine bal ance con fers atypi cal ity. This, how ever, is not borne out by the re sults. While 5-HT 2A re cep tors are read ily blocked at low dos ages of most atypi cal an tipsy chotic drugs (with the im por tant ex cep - tions of re moxi pride and amisul pride, nei ther of which is avail able for use in Can ada) the do s ages at which this hap pens are be low those needed to al le vi ate psy cho sis. In fact, the an tipsy chotic thresh old oc cu pancy of D 2 for an tipsy chotic ac tion re mains at about 65% for both typi cal and atypi cal an tipsy - chotic drugs, re gard less of whether 5-HT 2A re cep tors are blocked or not. At the same time, the an - tipsy chotic thresh old oc cu pancy of D2 for elic it ing EPS re mains at about 80% for both typi cal and atypi cal an tipsy chot ics, re gard less of the oc cu pancy of 5-HT 2A re cep tors. Rele vance: The fast- off-d 2 the ory, on the other hand, pre dicts which an tipsy chotic com pounds will or will not pro duce EPS and hy per pro lac ti ne mia and which com pounds pres ent a rela tively low risk for tar dive dyski ne sia. This the ory also ex plains why L- dopa psy cho sis re sponds to low atypi cal an tipsy chotic dos ages, and it sug gests vari ous in di vidu al ized treat ment strate gies. (Can J Psy chia try 2002;47:27 38) W Can J Psy chia try, Vol 47, No 1, Feb ru ary

2 The Ca na dian Jour nal of Psy chia try In Re view Key Words: typical antipsychotics, atypical antipsychotics, schizophrenia, dopamine receptors, 5-HT receptors Definition of Typical and Atypical Antipsychotic Drugs Traditional or typ i cal antipsychotics such as haloperidol and chlorpromazine, when used in clin i cally ef fec tive dos - ages, in duce el e vated levels of se rum prolactin, extrapyramidal signs and symp toms (EPS) and, af ter a pe riod of time, tardive dyskinesia (TD). How ever, antipsychotic drugs such as olanzapine, clozapine, quetiapine, and amisulpride are atyp i cal be cause, in con trast to the tra di - tional antipsychotics, they elicit low or neg li gi ble lev els of these un to ward side ef fects while still ef fec tively con trol ling psy chotic symp toms. Which Neuron Pathway Is Clinically Most Affected by Antipsychotic Drugs? Im me di ately af ter the clin i cal in tro duc tion of drugs for psy - cho sis (1), cli ni cians ob served that pa tients tak ing these med - i ca tions exhibited a Par kin son-like syndrome of tremor, akinesia, and rigidity (2). This drug-induced parkinsonism strongly sug gested that antipsychotic drugs were in ter fer ing with do pa mine path ways in the hu man brain, be cause Par kin - son s dis ease was known to be a dis ease of in suf fi cient do pa - mine neurotransmission. This clin i cal ob ser va tion gave birth to the do pa mine hy poth e sis of psy cho sis and antipsychotic drug ac tion (3). Al though it was sug gested that chlorpromazine and haloperidol blocked 5-hydroxytryptamine (serotonin) and monoaminergic (noradrenaline and do pa mine) receptors (4), it was not pos si ble at that time to con clude which of the 3 path ways was se lec tively af fected by antipsychotics. This is be cause the turn over of noradrenaline, se ro to nin (5-HT), and do pa mine were all si mul ta neously af fected by the antipsychotics (4,5). Andén and oth ers spec u lated that chlorpromazine and haloperidol re duce the elim i na tion rates of these me tab o lites of noradrenaline, 5-HT, and do pa mine (5). Al though Andén and oth ers (6) sub se quently found that antipsychotic drugs in vivo had a greater ef fect on do pa mine turn over than on noradrenaline turn over, di rect in vi tro ev i - dence for the se lec tive block ade of do pa mine re cep tors was found only later (7 9). The mul ti ple clin i cal and ad verse ef fects of var i ous antipsychotic drugs de pend on the com bi na tion of re cep tors oc cu pied, but the do pa mine path way is the pri mary com mon tar get for all antipsychotic drugs. More spe cif i cally, no drug has yet been iden ti fied with antipsychotic ac tion with out a sig nif i cant af fin ity for the D 2 re cep tor (10,11). Which Group of Dopamine Receptors Is Clinically Relevant for Antipsychotic Drug Action? There are 5 types of dopamine re cep tors in hu man beings (12,13). Types 1 and 5 are sim i lar in struc ture and drug sen si - tivity (14,15), and these 2 re cep tors are referred to as the D 1 -like group or class of re cep tors. Do pa mine re cep tor types 2, 3, and 4 are also similar in structure and are, there fore, grouped to gether as the D 2 -like group. Do pa mine re cep tors 2, 3 and 4, how ever, have sig nif i cantly dif fer ent sen si tiv i ties to antipsychotic drugs. Al though the D 1 -like re cep tors are of ten men tioned as a pri - mary tar get for antipsychotic drugs (16), 3 find ings in di cate that the D 1 -like re cep tors are not clin i cally rel e vant in the ther - a peu tic ac tion of these drugs. First, D 1 an tag o nists do not clin i - cally im prove psy chotic signs and symp toms (17 19). Sec ond, therapeutic main te nance dos ages of var i ous antipsychotic drugs occupy low or neg li gi ble lev els of D 1 re cep tors in the brains of pa tients with psy cho sis (20). For ex am ple, ther a peu - tic dos ages of haloperidol oc cupy less than 5% of the do pa - mine D 1 re cep tors in the brain putamen of schizophrenia pa tients (20). Al though ther a peu tic dos ages of some antipsychotic drugs, such as clozapine, oc cupy ap prox i mately 36% to 59% of brain do pa mine D 1 re cep tors (21), there is no cur rently known rea son to be lieve that these oc cu pied D 1 re - cep tors con trib ute to the unique prop er ties of clozapine. Third, for the D 1 do pa mine re cep tor, the bind ing con stants (that is, the dis so ci a tion con stants, also re ferred to as the in hi bi tion con - stants, or Ki val ues) of var i ous antipsychotic drugs (22) are very much higher than the concentrations of antipsychotic drugs found in the cerebrospinal fluid or in the plasma wa ter of pa tients (12,23,24). In other words, if the free con cen tra tions of antipsychotic drugs were as high as the val ues for the bind - ing con stants at D 1, the drugs would be toxic or lethal to patients. Of the 3 D 2 -like re cep tors, only the D 2 re cep tor it self is blocked by antipsychotic drugs in direct relation to their clin i cal antipsychotic potencies (8,9,25). Although this long-known re la tion is some times crit i cized as sim ply a re la tion be tween the D 2 -blocking con cen tra tions and the clin i cal dos ages at which EPS first ap pear, it is im por tant to note that the con cen - tra tions of antipsychotics which block D 2 re cep tors in the brain are pre cisely iden ti cal to the con cen tra tions found in the spi nal fluid or plasma wa ter (that is, cor rected for drug bind ing to the plasma pro teins) of pa tients whose psy chotic symp toms are suc cess fully con trolled by antipsychotics (Fig ure 1). Be cause it is known that the clin i cal ef fi cacy of antipsychotics is as so ci - ated with a block ade of 60% to 80% of D 2 re cep tors in the brain (26 28), the antipsychotic con cen tra tions shown on the or di - nate in Fig ure 1 are those needed to block 75% of D 2 re cep tors 30 W Can J Psy chia try, Vol 47, No 1, Feb ru ary 2002

3 Atypical Antipsychotics: Mechanism of Action Fig ure 1. The con cen tra tions of an tipsy chotic drugs that block do pa mine D2 re cep - tors in vi tro (us ing [ 3 H]ra clo pride) are iden ti cal to the con cen tra tions of an - tipsy chotic drugs that are found in the spi nal fluid or in the plasma wa ter (that is, cor rected for drug bind ing to the plasma pro teins) of pa tients be ing suc cess fully main tained on these drugs. The an tipsy chotic con cen tra - tions needed to block 75% of D2 re - cep tors in vi tro is shown (or di nate) be cause it is known that the clini cal ac tion of an tipsy chot ics is as so ci ated with a block of 60% to 80% of D2 re - cep tors. (See text for ad di tional de - tails.) in the pres ence of a phys i o log i cal con cen tra tion of do pa mine (see next sec tion). Endogenous Dopamine Raises the Antipsychotic Concentration Needed for D 2 Block Upon en try into the syn ap tic space, the antipsychotic drug must com pete with en dog e nous do pa mine for the re cep tor. Thus, the antipsychotic ther a peu tic con cen tra tion needed to block 50% of do pa mine re cep tors in the pres ence of do pa mine will be higher than that needed in the ab sence of do pa mine. This is in ac cor dance with the equa tion C50% = Ki [1+D/D 2 high], where D is the do pa mine con cen tra tion in the syn ap tic space and where D 2 high is the dis so ci a tion con stant of do pa - mine at the high-affinity state of the do pa mine D 2 re cep tor. The level of do pa mine in the syn ap tic space in hu mans is not known but, in the rat nu cleus accumbens, it is be tween 1 and 4 nm at rest, mo men tarily ris ing to 200 nm for the few mil li sec - onds it takes for a nerve im pulse to prop a gate (29). The do pa mine D 2 re cep tor can ex ist in ei ther a high-affinity state or a low-affinity state for do pa mine. The high-affinity state, D 2 high, is the phys i o log i cally func tional state (30). The dis so ci a tion con stant of do pa mine at D 2 high is 1.75 nm (see later). Hence, al though the con cen tra tion of do pa mine, D, in the hu - man syn ap tic space is not known, it ap pears that D is of the same or der of mag ni tude as the do pa mine Ki for D2 high. Hence, with this sin gle as sump tion that D is equiv a lent to D2 high, the above equa tion of C50% = Ki [1 + D / D2 high] re - duces to C50% is equiv a lent to 2 Ki. The frac tion, F, of D2 re cep tors oc cu pied by an antipsychotic at a con cen tra tion C is C / (C + Ki). Using this formula, it can be shown that the con cen tra tion of an antipsychotic drug needed to oc cupy 75% of the D2 re cep tors is about 3 times higher than that re quired to oc cupy 50% of the re cep tors. There fore, us ing the above equa tions, the antipsychotic con - cen tra tions to oc cupy 75% of D2 re cep tors in pa tients were cal cu lated and found to be vir tu ally iden ti cal to the ther a peu tic con cen tra tions of the antipsychotic drugs in the cerebrospinal fluid or in the plasma wa ter (that is, cor rected for drug bind ing to the plasma pro teins) of pa tients be ing suc cess fully main - tained on these med i ca tions. This is il lus trated in Fig ure 1. Fast-Off Theory of Atypical Antipsychotic Action: Atypicals Are Rapidly Released from D 2 Receptors As noted above, clin i cally ef fec tive dos ages of antipsychotic drugs oc cupy be tween 60% and 80% of brain do pa mine D2 re cep tors in pa tients, as mea sured by pos i tron emis sion to - mog ra phy (PET) or single photon emission tompography (SPET) in the hu man striatum (28,32 37,38 53). Clozapine and quetiapine, how ever, have con sis tently been ap par ent ex - cep tions. For ex am ple, in pa tients tak ing ther a peu ti cally ef - fective antipsychotic dosages of clozapine, this drug only occupies be tween 0% and ap prox i mately 50% of brain do pa - mine D2 re cep tors, as mea sured by var i ous radioligands us ing ei ther PET (26,28,31,34 36,54 58) or SPET (47 52,59). Because the atypical antipsychotics occupy many different types of re cep tors un der ther a peu tic con di tions, any ap par ent ex cep tions to the 60% to 80% rule of D2 oc cu pancy must be taken se ri ously. For ex am ple, the ap par ently low oc cu pancy of D2 by clozapine might sug gest that D2 is not the major antipsychotic tar get for clozapine (31,60). This is an im por - tant point be cause, if D2 is not the com mon target for all antipsychotic drugs, then the ex pla na tion for atypicality must W Can J Psy chia try, Vol 47, No 1, Feb ru ary

4 The Ca na dian Jour nal of Psy chia try In Re view Fig ure 2 Hu man cloned D2 re cep tors were equili brated with the tritium- labelled an tipsy chotic drug, af ter which a high con cen tra tion of ra clo pride or do pa - mine was used to dis place the an - tipsy chotic drug. The typi cal an tipsy chotic drugs chlor pro maz ine, ha loperi dol, and ra clo pride dis so ci - ated slowly over 30 min utes, while the atypi cal an tipsy chot ics dis so ci - ated rap idly in un der 60 sec onds. Ol - an zap ine and sert in dole had in ter me di ate rates of dis so cia tion. (See refs. 61,62 for de tails.) The fi nal con cen tra tion of ra clo pride was 100 mi cro mo lar, in con trast to that of 10 mi cro mo lar used ear lier (ref. 61), ex - plain ing why the 50% off set times for ra dio ac tive ha loperi dol, or ra dio ac tive sert in dole, were lower than the pre vi - ously pub lished data (ref. 61). Fig ure 3 Positron emission tomography imaging (using [ 11 C]raclopride) reveals that the human brain (striatum) occupancy of D2 by quetiapine and clozapine rapidly falls off within 24 hours, in contrast to that for ha loperi dol, which maintains its D 2 occupancy constant over 24 hours (adapted from refs. 38,53,63). be found else where, per haps in the 5-HT sys tem or in the bal - ance be tween 5-HT and do pa mine. How ever, the ap par ently low oc cu pancy of D 2 by clozapine and quetiapine is readily explained by the fact that these 2 antipsychotics rap idly dis so ci ate from the do pa mine D2 re - cep tor (27). This also holds for remoxipride and amisulpride, 2 atyp i cal drugs not used clin i cally in Can ada. For ex am ple, hu man-cloned do pa mine D 2 re cep tors re lease [ 3 H]clozapine, [ 3 H]quetiapine, [ 3 H]remoxipride, and [ 3 H]amisulpride at least 100 times faster than they release [ 3 H]haloperidol or [ 3 H]chlorpromazine (27,61,62). Fig ure 2 shows the rapid re lease of [ 3 H]clozapine, [ 3 H]quetiapine, [ 3 H]remoxipride, and [ 3 H]amisulpride from human-cloned do pa mine D 2 re cep tors; the slow re lease of [ 3 H]raclopride, [ 3 H]haloperidol, and[ 3 H]chlorpromazine; and the in ter me di ate re lease rates from these re cep tors for [ 3 H]olanzapine and [ 3 H]sertindole. These in vi tro data match those found clin i cally for clozapine, quetiapine, and haloperidol in schizo phre nia pa tients and healthy vol un teers. This is shown in Fig ure 3, where it has been found by PET (using [ 11 C]raclopride) that the human brain (striatum) oc cu pancy of D 2 by quetiapine and clozapine rap idly falls off within 24 hours, in contrast to that for haloperidol, which main tains its D2 oc cu pancy con stant over 24 hours (38,53,63). Thus, the rapid re lease of clozapine and quetiapine from do pa - mine D 2 re cep tors and their re place ment by en dog e nous do pa - mine would readily ac count for the low D 2 re cep tor oc cu pancy shown by these atyp i cal antipsychotics. It is im por tant to em pha size that the rapid re lease of clozapine and quetiapine is a molecular event which oc curs quickly, 32 W Can J Psy chia try, Vol 47, No 1, Feb ru ary 2002

5 Atypical Antipsychotics: Mechanism of Action re gard less of the clin i cal dos age used. In other words, even though high dos ages of clozapine and quetiapine may be used, these drugs con tinue to go on and off the D 2 re cep tor rap idly, al low ing ex ten sive and fre quent ac cess of en dog e nous do pa - mine to the receptor. Hence, it ap pears that some antipsychotics, such as clozapine and quetiapine, oc cupy D 2 re cep tors only tran siently through - out the day. As just men tioned, PET im ag ing of pa tients with schizophrenia reveals that the D 2 re cep tor occupancies by clozapine and quetiapine wear off quickly af ter an oral dos - age, and pa tients may show no oc cu pancy what so ever within 48 hours of the last dose, in con trast to typ i cal antipsychotics, which may con tinue to oc cupy D 2 receptors for days. This may ex plain why psy chotic re lapses of pa tients on clozapine and quetiapine oc cur soon af ter with drawal of the antipsychotic (64,65, re viewed in 27) much ear lier than af - ter with drawal of con ven tional antipsychotic drugs such as haloperidol or chlorpromazine. Clinical and Basic Implications of the Fast-Off Theory of Atypical Antipsychotic Action As out lined above, the fast-off the ory of atyp i cal antipsychotic ac tion is that the atypicals have low af fin i ties for the do pa mine D 2 re cep tor, and are loosely bound to, and rap - idly re leased from, these re cep tors. A crit i cal as pect of the the - ory is that the atyp i cal antipsychotics bind more loosely to D 2 than does do pa mine it self, while the tra di tional, typ i cal antipsychotics bind more tightly than do pa mine. These data are sum ma rized in Ta ble 1 and Fig ure 4A. Fig ure 4A il lus trates a gen eral de mar ca tion be tween typicals and atypicals. That is, the typ i cal antipsychotics have K val ues lower than that for do pa mine (at the high-affinity state of the D 2 re cep tor), while the atypicals have K val ues higher than that for do pa mine. Al though risperidone ap pears to be an ex - ception to this gen er al iza tion, risperidone is the weak est atyp i cal antipsychotic, elic it ing dos age-dependent EPS in 60% to 70% of pa tients tak ing 6 mg or more daily, a dos age that may be in suf fi cient for clin i cal ef fi cacy (66). Clearly, the sep a ra tion be tween typicals and atypicals in Fig - ure 4A is not sharp and pre cise, be cause antipsychotic drugs with K val ues be tween 2 nm and 10 nm (in clud ing molindone and loxapine) of ten re veal dose-dependent EPS. Thus, the de - mar ca tion be tween typ i cal and atyp i cal antipsychotics is not a sharp di vide but rather a con tin u ous one. Antipsychotics be - come in creas ingly more atyp i cal as their bind ing to the D 2 re - cep tor be comes more loose and they are re leased more quickly. Al though all atyp i cal antipsychotics have loose bind - ing, with dis so ci a tion con stants looser than 1.8 nm/l, they can still elicit dos age-dependent parkinsonism. For ex am ple, olanzapine, with a dis so ci a tion con stant of 5.1 nm, is known to be as so ci ated with a dose-dependent in ci dence of EPS in some pa tients and es pe cially at higher dos ages. If the bind ing is ex tremely loose, as with clozapine, remoxipride, quetiapine, and melperone, es sen tially no EPS oc curs (al - though ex qui sitely sen si tive pa tients do ex ist who will ex hibit EPS even with these drugs). Drugs that are too loose or have far too low an af fin ity for D 2 re cep tors cease to ex hibit any antipsychotic ac tiv ity at all. More over, al though the de gree of oc cu pancy of atypicals at D 2 re cep tors has a di rect in flu ence on EPS, the po tent anticholinergic ac tion of olanzapine and clozapine pro vides an ad di tional anti-eps mech a nism. It is because of its anticholinergic properties, for in stance, that thioridazine use is rel a tively free of EPS. Ta ble 2 sum ma rizes a few clin i cal dis tinc tions be tween the typ i cal antipsychotics, which are tightly bound to D 2, and the atyp i cal antipsychotics, which are loosely bound to D 2. The required antipsychotic dos age (in mg) will be low for tightly bound drugs but high for loosely bound drugs. The typicals, be ing tightly bound to D 2, will elicit EPS and el e vated prolactin, while the atypicals, be ing loosely bound and rap idly re leased from D2, will not elicit these side ef fects, or will at least elicit them to a mark edly lesser ex tent. Finally, be cause the typicals re main at tached to D2 and readily ac cu mu late in brain tis sue, they will even tu ally lead to TD (67). The atypicals, however, are much less fat-soluble, and because they are readily re leased from D 2 and from the brain tis sue, the risk of caus ing TD is much re duced or per haps ab sent. L-dopa Psychosis: Fast-Off-D 2 Theory Predicts Low Dosage of Atypical Antipsychotics The treat ment of pa tients with psy cho sis in Par kin son s dis - ease (as a con se quence of L-dopa treat ment) is best done with a very loose binding antipsychotic, such as clozapine or quetiapine, to al low for the low level of do pa mine neurotransmission that is re quired for nor mal mo tor func tion - ing to con tinue. Par kin son pa tients are do pa mine-depleted, so it is very im por tant not to block the lit tle do pa mine func tion that re mains. The hy poth e sis is that atyp i cal antipsychotic ac - tion (that is, low EPS and nor mal prolactin) oc curs when en - dog e nous do pa mine is able to dis place a loosely bound antipsychotic. This ac cords with the ob ser va tion that low dos - ages of atyp i cal antipsychotics are useful for Par kin son patients. It is well known in neu rol ogy that L-dopa psy cho sis in a pa - tient with Par kin son s dis ease is best treated with a dos age of clozapine that is about 10% the dos age nor mally used for psy - cho sis in schizo phre nia. The fast-off-d 2 hy poth e sis readily and quantitatively pre dicts this. As pre sented above, the antipsychotic dosage needed to oc cupy D 2 re cep tors is W Can J Psy chia try, Vol 47, No 1, Feb ru ary

6 The Ca na dian Jour nal of Psy chia try In Re view Table 1. Antipsychotic dissociation constants at doapmine and serotonin receptors K value, nm K value, nm Ra tio: K value, nm Do pa mine D 2 5-HT 2A D 2 K 5-HT 2A re cep tor a re cep tor b 5-HT 2A K receptor c M100,907 (not antipsychotic) , Melperone (atypical) Perlapine (atypical) Quetiapine (atypical) 122 d Remoxipride (atypical) no ef fect Clozapine (atypical) 63 e 3.7 m Amoxapine (atypical) Sulpiride-S (atypical) 9.9 no ef fect Loxapine (typical) Iloperidone (atypical) Olanzapine (atypical) 5.1 f 2.5 m Molindone (typical) no ef fect Ziprasidone (atypical) Sertindole (atypical) 2.3 g 0.28 m Amisulpride-S (atypical) 1.8 h Dopamine at D 2high 1.75 n Raclopride (typical) 1.7 i E-04 Prochlorperazine (typical) 1.7 Moperone (typical) Pimozide (typical) Trifluperazine (typical) Risperidone (atypical?) Thioridazine (typical) Chlorpromazine (typical) 0.99 j Chlorprothixene (typical) 0.7 Haloperidol (typical) 0.55 k Fluphenazine (typical) Droperidol (typical) 0.54 Flupentixol-cis (typical) Perphenazine (typical) 0.27 Thiothixene-cis (typical) Butaclamol-(+) (typical) Spiperone (typical) 0.04 l Epidepride (typical) Nemonapride (typical) a Human cloned do pa mine D 2short re cep tor in GH4C1 cells (ob tained from Al le lix., Mis sis sauga, or from Bio sig nal Inc., Mont real); D 2 in CHO cells were oc ca sion ally used. K val ues were ob tained [2 to 25 meas ure ments] us ing 2 nm [ 3 H]ra clo pride which had a Kd value of 1.9 nm. b Human cloned 5-HT2A re cep tors in HEK293 cells (ob tained from Al le lix); K values were obtained [2 to 8 measurements] using 0.5 nm [ 3 H]ketanserin which had a Kd value of 0.55 nm. c K values of antipsychotics causing inverse agonism at 5-HT2A receptors; data from ref 79. d Average of quetiapine Ki (140 nm) and [ 3 H]quetiapine Kd (104 nm); 13 Ci/mmol. e Average of clozapine Ki (75 nm) and [ 3 H]clozapine Kd (51 nm); 84 Ci/mmol. f Average of olanzapine Ki (7.4 nm) and [ 3 H]olanzapine Kd (2.7 nm); 81 Ci/mmol. g Average of sertindole Ki (1.9 nm) and [ 3 H]sertindole Kd (2.6 nm); 47 Ci/mmol. h Average of amisulpride Ki (1.8 nm) and [ 3 H]amisulpride Kd (1.8 nm); 84 Ci/mmol. i Average of raclopride Ki (1.5 nm) and [ 3 H]raclopride Kd (1.9 nm); Ci/mmol. j Average of chlorpromazine Ki (1.2 nm) and [ 3 H]chlopromazine Kd (0.77 nm); 27 Ci/mmol. k Average of haloperidol Ki (0.7 nm) and [ 3 H]haloperidol Kd (0.4 nm); 8.0 Ci/mmol. l Average of spiperone Ki (0.018 nm) and [ 3 H]spiperone Kd (0.065 nm); Ci/mmol. m Average of antipsychotic Ki (vs [ 3 H]ketanserin; 81 Ci/mmol) and Kd using [ 3 H]antipsychotic. n Average of dopamine Ki (2.1 nm) (9 measurements) and [ 3 H]dopamine Kd (1.3 nm) at D2 High; 54 Ci/mmol. 34 W Can J Psy chia try, Vol 47, No 1, Feb ru ary 2002

7 Atypical Antipsychotics: Mechanism of Action Fig ure 4a 4b, and 4c. Com par ing 3 theo ries for atypi cal an tipsy chotic ac tion. Typi cal an tipsy chot ics are col o ured in pink, atypi cals in green. Risperi done is a weak atypi cal, and is col oured half- green- half pink. A: The fast- off- D 2 the ory pro poses that typi cal an tipsy chot ics bind more tightly than do pa mine to the do pa mine D 2 re cep tor (in its func tional high- affinity state), with dis so cia tion con stants lower than that for do pa mine, while the atypi cals bind more loosely than do pa mine to the do pa mine D 2 re cep tor, with dis so cia tion con stants higher than that for do pa mine. Out of 31 an tipsy chot ics, there are 2 or 3 ap par ent ex cep tions to this rule. Drugs with K val ues be tween 2 and 10 nm cause dose- dependent ex tra py ra mi dal signs. B: The do pa mine-se ro tonin an tago n ism the ory gen er ally pre dicts a sepa - ra tion be tween typi cals and atypi cals, ex cept that out of 20 an tipsy chot ics there are 3 or 4 ap par ent ex cep tions to this the ory. Re - moxi pride is an im por tant ex cep tion. C: The the ory which pre dicts that an tipsy chot ics stimu late 5-HT2A re cep tors by in verse ago nism has many ex cep tions, in clud ing M100,907 which has no an tipsy chotic ac tion. W Can J Psy chia try, Vol 47, No 1, Feb ru ary

8 The Ca na dian Jour nal of Psy chia try In Re view Table 2. Aspects of tight and loose antipsychotic binding at dopamine D 2 receptors Tight Loose Dosage Low High Extrapyramidal symptoms (EPS) Yes No Prolactin High Normal Tardive dyskinesia High risk Low risk pro por tional to K [1 + D / Dhigh], where K is the dis so ci a - tion con stant of the antipsychotic, D is the con cen tra tion of do pa mine in the syn ap tic space dur ing the mo men tary nerve im pulse (equiv a lent to 200 nm, ref. 29), and where Dhigh is the dissociation constant of do pa mine at the high-affinity state of D 2 (equiv a lent to 1.75 nm; Ta ble 1). In Par kin son s dis ease, where 90% to 95% of the do pa mine con tent is ab sent, the value for D would be equiv a lent to 20 nm. Ac cord ingly, the antipsychotic dos age for L-dopa psy cho sis will be lower than that for schizo phre nia psy cho sis by a fac tor of (1 + D / Dhigh) nor mal / (1 + D / Dhigh) Par kin son, or ( / 1.75) / ( / 1.75), or ten fold. Thus, while a daily dos age of 500 mg clozapine might be suitable for treating schizophrenia psy cho sis, a dos age of 50 mg (or less) would be more than ad - e quate to treat L-dopa psy cho sis. It is im por tant to note that this cal cu la tion best holds for com pe ti tion be tween en dog e - nous do pa mine and a loosely bound antipsychotic. A tightly bound antipsychotic such as haloperidol would not readily per mit en dog e nous do pa mine to re place it com pet i tively. Test of the Fast-Off-D 2 Hypothesis Using Clozapine and Isoclozapine As re ported by Kapur and oth ers (68), the sin gle most pow er - ful pre dic tor of atypicality is the low af fin ity to, and fast dis - so ci a tion from, the D 2 re cep tor not high af fin ity to any other re cep tor. This hy poth e sis is sup ported by their find ings that clozapine and isoclozapine have iden ti cal po ten cies on many cloned re cep tors (in clud ing muscarinic M 1, do pa mine D 1, do - pa mine D 4, 5-HT 1A, and 5-HT 2A re cep tors) but dif fer five fold in their po tency only on D 2 re cep tors. Thus, in sev eral tests of atypicality (for ex am ple, early activation of certain genes, cat a lepsy in an i mals, and prolactin el e va tion), clozapine be - haves like an atyp i cal antipsychotic. Isoclozapine, how ever, be haves like a con ven tional antipsychotic. Do Antipsychotics Elicit Atypical Action by Blocking 5-HT Receptors? In ad di tion to block ing do pa mine re cep tors, the new atyp i cal antipsychotic drugs also block 5-HT re cep tors. Al though it has often been sug gested that the blockade of 5-HT 2A re cep tors may al le vi ate the parkinsonism caused by D 2 block - ade (69,70), most data do not sup port this prin ci ple. Remoxipride Is an Im por tant Ex cep tion Remoxipride is a highly ef fec tive atyp i cal antipsychotic drug (not used in Can ada) with no EPS and no hyperprolactinemia, yet it does not block 5-HT re cep tors. 5-HT Block ade En hances Cat a lepsy Se lec tive 5-HT2A re cep tor block ade with the drug M100,907 mark edly en hances, in stead of re duc ing, the cat a lepsy (cat a - lepsy in an i mals = EPS in hu mans) ob served with submaximal dos ages of the D 2 block by raclopride (71). No Do pa mine 5-HT Cor re la tion to Cat a lep tic Dos ages There is no correlation be tween the catalep tic dosages of neuroleptics and the ra tio of the antipsychotic dis so ci a tion con stants at D 2 and at 5-HT 2A re cep tors (72,73). No Sharp Sep a ra tion of Typicals and Atypicals Using the ra tio of antipsychotic dissociation con stants ob - tained in our lab o ra tory on hu man-cloned D 2 and 5-HT2A re - cep tors (Ta ble 1), the de mar ca tion be tween typ i cal and atyp i cal antipsychotics shown in Fig ure 4B is not sharp and is less clear than that found in Fig ure 4A. For ex am ple, of the 20 antipsychotics in Fig ure 4B, there are 3 to 4 ap par ent ex cep - tions to the sep a ra tion of typicals and atypicals. This com pares to 2 or 3 ap par ent ex cep tions out of the 31 antipsychotics shown in Fig ure 4A. (Ex cep tions do not nec es sar ily kill a the - ory, but ex pla na tions for the ex cep tions have to be found). No Al le vi a tion of Extrapyramidal Signs A high de gree of 5-HT 2A receptor oc cu pancy (95%) by risperidone (6 mg daily) does not pre vent EPS in 6 out of 7 pa - tients (74,75). 5-HT Block Does Not Change D 2 Oc cu pancies Re quired Using [ 11 C]raclopride for imaging brain D 2 re cep tors and [ 11 C]setoperone for im ag ing brain 5-HT 2A re cep tors, Kapur and oth ers found that the high oc cu pancy of 5-HT 2A re cep tors by olanzapine or by risperidone did not al ter ei ther the D 2 oc - cu pancy re quired for the antipsychotic ef fect or the D 2 oc cu - pancy at which EPS occur (28). The threshold dos ages for antipsychotic ac tion con sis tently oc cupy 65% of brain D 2 re - cep tors in pa tients, and the thresh old dos ages for EPS con sis - tently oc cupy 80% of brain D 2 re cep tors in pa tients, whether or not the 5-HT 2A re cep tors are oc cu pied. As il lus trated in Fig ure 5, the results showed that the oc cu pancy of D 2 re cep tors in first-episode schizo phre nia pa tients was 65% for antipsychotic thresh old dos ages of haloperidol (1.5 to 2.1 mg daily) and olanzapine (7.5 to 10 mg daily), de spite the neg li gi - ble oc cu pancy of 5-HT 2A re cep tors by haloperidol or the very high oc cu pancy, ex ceed ing 95%, by olanzapine. It is im por tant to note that while first-episode pa tients may not be typ i cal of 36 W Can J Psy chia try, Vol 47, No 1, Feb ru ary 2002

9 Atypical Antipsychotics: Mechanism of Action Fig ure 5. The thresh old doses for an tipsy - chotic ac tion con sis tently oc cupy 65% of brain D2 re cep tors in pa - tients, and the thresh old doses for ex tra py ra mi dal signs con sis tently oc cupy 80% of brain D 2 re cep tors in pa tients, whether or not the 5-HT 2A re cep tors are oc cu pied. As shown here, the oc cu pancy of D2 re cep tors in first- episode schizo - phre nia pa tients was 65% for an - tipsy chotic thresh old doses of ha loperi dol (1.5 to 2.1 mg daily) and ol an zap ine (7.5 to 10 mg daily), de spite the neg li gi ble oc cu pancy of 5-HT 2A re cep tors by ha loperi dol or the very high oc cu pancy, ex ceed ing 95%, by ol an zap ine. (Adapted from ref. 28). pa tients with chronic schizo phre nia, stud ies with the first-episode pa tients are ex ceed ingly im por tant in working out mechanisms of antipsychotic action, be cause they have had not pre vi ous ex po sure to drugs. It is not clear what clin i cal ben e fit, if any, is pro vided by the block ade of 5-HT re cep tors. Al though low dos ages of cyproheptadine have been used (64) to block 5-HT 2A re cep tors and sup ple ment antipsychotic ad - min is tra tion, it should be noted that cyproheptadine has a D 2 block ing ac tion. It has a K of 24 nm at D 2 re cep tors, com pared with 63 nm for clozapine and 21 nm for amoxapine (Ta ble 1). Amoxapine, though mar keted as an an ti de pres sant, has antipsychotic prop er ties. Amisulpride Is an Im por tant Ex cep tion Amisulpride (used in Eu rope) is a highly ef fec tive antipsychotic that is atyp i cal and does not oc cupy any 5-HT 2A re cep tors in hu mans at dos ages up to 1200 mg daily (76). Chlorpromazine Blocks 5-HT but Elicits EPS Chlorpromazine, the first typical antipsychotic, blocks 65% of 5-HT 2A receptors at 500 mg daily. This high level of 5-HT 2A block... sug gests that the dis tinct clin i cal pro files of chlorpromazine and clozapine are un re lated to 5-HT 2A re cep - tor block ade (76). 5-HT Block Not Needed for Antipsychotic Ac tion It has also been stated that the block of 5-HT 2A re cep tors is not a pre req ui site for the antipsychotic ef fect (74,75). In fact, full block of 5-HT 2A re cep tors oc curs at subtherapeutic dos - ages of risperidone, olanzapine, and clozapine, in di cat ing that 5-HT 2A block has lit tle or no antipsychotic ac tion. Do Antipsychotics Elicit Atypical Action by Stimulating 5-HT Receptors? Al though it has long been known that the stimulation of 5-HT 1A re cep tors in an i mals can al le vi ate cat a lepsy caused by D 2 block ade (77,78), there do not ap pear to be any antipsychotics that have this 5-HT 1A -stim u lat ing ac tion com - bined with D 2 -block ing ac tion. It has re cently been pro posed that the stim u la tion of 5-HT 2A re cep tors by an in verse ac tion is an im por tant con tri bu tion to atyp i cal antipsychotic ac tion (79). This is il lus trated in Fig ure 4C, where the in verse stim u lat ing po ten cies of antipsychotics on 5-HT 2A re cep tors are shown (from 79). How ever, be cause a few im por tant atyp i cal antipsychotics (in clud ing remoxipride and sulpiride) have no such stim u lat ing ac tion, it is un likely that this fea ture con trib utes to atyp i cal antipsychotic ac tion. The data in Fig ure 4C do not re veal any clear de mar ca tion be - tween typicals and atypicals. Finally, although the authors (79) pro pose that M100,907 has the de sired stim u lat ing ac - tion, this com pound has shown no antipsychotic ac tiv ity in humans. The Future Be cause brain im ag ing in di cates that the tra di tional antipsychotics re main at tached to do pa mine D 2 re cep tors for at least 1 or 2 days, there is no ra tio nal need to medicate schizo phre nia pa tients daily with typ i cal antipsychotics. This rea son ing has led to a new reg i men of ad min is ter ing antipsychotics by ex tended dos ing, wherein the pa tient re - ceives a typ i cal antipsychotic once ev ery 3rd or 4th day (80). Such a pro ce dure, of course, could not be used with atyp i cal antipsychotics be cause of their loose bind ing to D 2 and sub se - quent risk of re lapse. W Can J Psy chia try, Vol 47, No 1, Feb ru ary

10 The Ca na dian Jour nal of Psy chia try In Re view Cli ni cians can now ap ply this knowl edge to the treat ment of individual patients. Atyp i cal agents, be ing newer and still pro tected by pat ent, are much more ex pen sive than the older drugs. The fact that they do not elicit EPS and do not el e vate prolactin lev els does not mean that they are free of se ri ous side ef fects. One could ar gue that the side ef fects as so ci ated with some of the atyp i cal drugs (for ex am ple, agran u lo cy to - sis, obe sity, di a be tes, ophthalmological prob lems, car dio vas - cu lar prob lems, sexual prob lems, ob ses sive com pul sive symp toms, con vul sions, and in som nia) are more se ri ous than EPS, high prolactin, and even TD. Low-dose, ex - tended-dosing reg i mens of typ i cal drugs may be best suited for spe cific pa tients. Pa tients known to be nonadherent to reg - u lar med i ca tion may do better on those drugs that are more tightly bound to the D 2 re cep tor, where risk of re lapse through a short pe riod of non com pli ance is re duced. Con versely, pa - tients with a his tory of neuroleptic ma lig nant syn drome are best treated with drugs that are readily dis placed, so that, should the syn drome re turn, the drug is quickly out of their brain. In pa tients with psy cho sis, high stress lev els, ac com pa - nied by high en dog e nous do pa mine re lease, will necessitate higher dos ages of the antipsychotic drug. Pe riods of low stress will require lower dosages. Pa tients with psy cho sis who may tem po rarily ben e fit from high prolactin lev els (for ex am ple, those who do not want to con ceive or, con versely, postpartum women whose milk is in suf fi cient for breast feed - ing) may pref er en tially be pre scribed typ i cal antipsychotics. On the other hand, typ i cal antipsychotics should be dis con tin - ued in those with be gin ning signs of TD and the newer drugs prescribed in stead. Knowing how drugs work greatly ex - pands the cli ni cian s rep er toire of strat e gies, al low ing op ti mi - za tion of drug reg i mens for in di vid u al ized treat ment. Ac knowl edge ments I thank Dr S Kapur (Cen tre for Ad dic tion and Men tal Health, To - ronto) for his ex cel lent and con tin u ing col lab o ra tion on im por tant as pects of this re search. This work was sup ported by the Ca na dian In sti tutes of Health Re search, the On tario Men tal Health Foun da - tion, the Na tional Al li ance for Re search on Schizo phre nia and De - pres sion, and the US Na tional In sti tute on Drug Abuse. Manuscript received and accepted December Anne and Max Tanenbaum Chair of Neu ro sci ence, Pro fes sor, De part ments of Phar ma col ogy and Psy chi a try, Uni ver sity of To ronto, To ronto, On tario. Ad dress for cor re spon dence : Dr P Seeman, De part ment of Phar ma col ogy, Med i cal Sci ence Build ing, Room 4344, Uni ver sity of To ronto, To ronto, ON M5S 1A8 philip.seeman@utoronto.ca References 1. De lay J, Deniker P, Harl J-M. Traitement des états d excitation et d agitation par une méthode médicamenteuse dérivée de l hibernothérapie. Ann Méd Psychologie 1952;110 (Part 2): Haase HJ, Janssen PAJ. The ac tion of neuroleptic drugs: a psy chi at ric, neu ro - logic and phar ma co log i cal in ves ti ga tion. Chi cago: Year Book Med i cal Publ; Van Rossum J The sig nif i cance of do pa mine-receptor block ade for the ac tion of neuroleptic drugs. In: Brill H, Cole J, Deniker P, Hippius H, Bradley PB, ed i tors. Neuropsychopharmacology, Pro ceed ings 5th Collegium Internationale Neuropsychopharmacologicum. Am ster dam: Excerpta Medica; p Carlsson A, Lindqvist M. Ef fect of chlorpromazine or haloperidol on for ma tion of 3-methoxytyramine and normetanephrine in mouse brain. Acta Pharmacol Toxicol 1963;20: Andén N-E, Roos B-E, Werdinius B. Ef fects of chlorpromazine, haloperidol and re ser pine on the lev els of phe no lic ac ids in rab bit cor pus striatum. Life Sci 1964;3: Andén N-E, Butcher SG, Corrodi H, Fuxe K, Ungerstedt U. Re cep tor ac tiv ity and turn over of do pa mine and noradrenaline af ter neuroleptics. Eur J Pharmacol 1970;11: Seeman P, Wong M, Lee T. Do pa mine re cep tor-block and nigral fi ber-impulse block ade by ma jor tran quil iz ers. Fed Proc 1974;33: Seeman P, Chau-Wong M, Tedesco J, Wong K. Brain re cep tors for antipsychotic drugs and do pa mine: di rect bind ing as says. Proc Natl Acad Sci USA 1975;72: Seeman P, Lee T, Chau-Wong M, Wong K. Antipsychotic drug doses and neuroleptic/do pa mine re cep tors. Na ture 1976;261: Pickar D. Pros pects for the pharmacotherapy of schizo phre nia. Lan cet 1995;345: Su T-P, Malhotra AK, Hadd K, Breier A, Pickar D. D2 do pa mine re cep tor oc cu - pancy: a cross over com par i son of risperidone with clozapine ther apy in schizo - phrenic pa tients. Arch Gen Psy chi a try 1997;54: Seeman P. Do pa mine re cep tor se quences. Ther a peu tic lev els of neuroleptics oc - cupy D2, clozapine oc cu pies D4. Neuropsychopharmacology 1992;7: Seeman P, Corbett R, Nam D, Van Tol HHM. Do pa mine and se ro to nin re cep tors: amino acid se quences, and clin i cal role in neuroleptic parkinsonism. Jpn J Pharmacol 1996;71: Sunahara RK, Niznik HB, Weiner DM, Stormann TM, Brann MR, Ken nedy JL, and oth ers. Hu man do pa mine D1 re cep tor en coded by an intronless gene on chro - mo some 5. Na ture 1990;347: Sunahara RK, Guan H-C, O Dowd BF, Seeman P, Laurier LG, Ng GYK, and oth - ers. Clon ing of the gene for a hu man do pa mine D5 re cep tor with higher af fin ity for do pa mine than D1. Na ture 1991;350: Lidow MS, Wil liams GV, Goldman-Rakic PS. The ce re bral cor tex: a case for a com mon site of ac tion of antipsychotic drugs. Trends Pharmacol Sci 1998;19: Karlson P, Smith L, Farde L, Härnryd C, Sedvall G, Wiesel F-A. Lack of ap par - ent antipsychotic ef fect of the D1-dopamine re cep tor an tag o nist SCH39166 in acutely ill schizo phrenic pa tients. Psychopharmacology 1995;121: Den Boer JA, van Megen HJGM, Fleischacker WW, Louwerens JW, Slaap BR, Westenberg HGM, and oth ers. Dif fer en tial ef fects of the D1-DA re cep tor an tag o - nist SCH39166 on pos i tive and neg a tive symp toms of schizo phre nia. Psychopharmacology 1995;121: De Beaurepaire R, Labelle A, Naber D, Jones BD, Barnes TRE. An open trial of the D1 an tag o nist SCH39166 in six cases of acute psy chotic states. Psychopharmacology 1995;121: Farde L, Nordström A-L. PET anal y sis in di cates atyp i cal cen tral do pa mine re cep - tor oc cu pancy in clozapine-treated pa tients. Br J Psy chi a try 1992;160 (Suppl 17): Nordström A-L, Farde L, Nyberg S, Karlsson P, Halldin C, Sedvall G. D1, D2, and 5-HT2 re cep tor oc cu pancy in re la tion to clozapine se rum con cen tra tion: A PET study of schizo phrenic pa tients. Am J Psy chi a try 1995;152: Seeman P, Niznik HB. Do pa mine D1 re cep tor phar ma col ogy. ISI At las of Sci Phar ma col ogy 1988;2: Seeman P. Do pa mine re cep tors and the do pa mine hy poth e sis of schizo phre nia. Syn apse 1987;1: Seeman P, Tallerico T. Antipsychotic drugs which elicit lit tle or no Parkinsonism bind more loosely than do pa mine to brain D2 re cep tors, yet oc cupy high lev els of these re cep tors. Mol Psy chi a try 1998;3: Creese I, Burt DR, Snyder SH. Do pa mine re cep tor bind ing pre dicts clin i cal and phar ma co log i cal po ten cies of antischizophrenic drugs. Sci ence 1976;192: Farde L, Nordström A-L, Wiesel F-A, Pauli S, Halldin C, Sedvall G. Pos i tron emis sion tomographic anal y sis of cen tral D1 and D2 do pa mine re cep tor oc cu - pancy in pa tients treated with clas si cal neuroleptics and clozapine. Re la tion to extrapyramidal side ef fects. Arch Gen Psy chi a try 1992;49: Seeman P, Tallerico T. Rapid re lease of antipsychotic drugs from do pa mine D2 re cep tors: an ex pla na tion for low re cep tor oc cu pancy and early clin i cal re lapse upon drug with drawal of clozapine or quetiapine. Am J Psy chi a try 1999;156: Kapur S, Zipursky RB, Remington G. Com par i son of the 5-HT2 and D2 re cep tor occupancy of clozapine, risperidone, and olanzapine in schizo phre nia: clin i cal and the o ret i cal im pli ca tions. Am J Psy chi a try 1999;156: Kawagoe KT, Garris PA, Wiedemann DJ, Wightman RM. Reg u la tion of tran sient do pa mine con cen tra tion gra di ents in the microenvironment sur round ing nerve ter - mi nals in the rat striatum. Neu ro sci ence 1992;51: George SR, Watanabe M, Di Paolo T, Falardeau P, Labrie F, Seeman P. The func tional state of the do pa mine re cep tor in the an te rior pi tu itary is in the high-affinity form. En do cri nol ogy 1985;117: W Can J Psy chia try, Vol 47, No 1, Feb ru ary 2002

11 Atypical Antipsychotics: Mechanism of Action 31. Farde L, Hall H. Pos i tron emis sion to mog ra phy ex am i na tion of chem i cal trans - mis sion in the liv ing hu man brain. Arzneim-Forsch 1992;42: Nyberg S, Nordström A-L, Halldin C, Farde L. Pos i tron emis sion to mog ra phy stud ies on D2 do pa mine re cep tor oc cu pancy and plasma antipsychotic drug lev - els in man. Int Clin Psychopharmacol 1995;10 (Suppl 3): Kapur S, Remington G, Zipursky RB, Wil son AA, Houle S. The D2 do pa mine re cep tor oc cu pancy of risperidone and its re la tion ship to extrapyramidal symp - toms: a PET study. Life Sci 1995;57:PL103 PL Kapur S, Remington G, Jones C, Wil son A, DaSilva J, Houle S, Zipursky RB. High lev els of do pa mine D2 re cep tor oc cu pancy with low-dose haloperidol treat - ment: a PET study. Am J Psy chi a try 1996;153: Kapur S, Zipursky RB, Jones C, Remington GJ, Wil son AA, DaSilva J, and oth - ers. The D2 re cep tor oc cu pancy pro file of loxapine de ter mined us ing PET. Neuropsychopharmacology 1996;15: Kapur S, Zipursky RB, Remington G, Jones C, DaSilva J, Wil son AA, and oth - ers. 5-HT2 and D2 re cep tor oc cu pancy of olanzapine in schizo phre nia: a PET in - ves ti ga tion. Am J Psy chi a try 1998;155: Knable MB, Heinz A, Coppola R, Gorey J, Weinberger DR. IBZM SPECT mea - sure ment of D2 re cep tor oc cu pancy by haloperidol and risperidone. Biol Psy chi a - try 1996;39: Farde L, Wiesel FA, Nordström A-L, Sedvall G. D1 and D2-dopamine re cep tor oc cu pancy dur ing treat ment with con ven tional and atyp i cal neuroleptics. Psychopharmacology 1989;99:S28 S Tauscher J, Küfferle B, Barnas C, Asenbaum S, Brücke T, Kasper S. Do pa - mine-2 re cep tor im ag ing in psy chotic pa tients treated with quetiapine, clozapine, risperidone and haloperidol. Biol Psy chi a try 1997;42:15S. 40. Küfferle B, Tauscher J, Asenbaum S, Vesely C, Podreka I, BrückeT, and oth ers. IBZM SPECT im ag ing of striatal do pa mine-2 re cep tors in psy chotic pa tients treated with the novel antipsychotic sub stance quetiapine in com par i son to clozapine and haloperidol. Psychopharmacology 1997;133: Nordström A-L, Farde L, Wiesel F-A, Forslund K, Pauli S, Halldin C, and oth - ers. Cen tral D2-dopamine re cep tor oc cu pancy in re la tion to antipsychotic drug ef fects: a dou ble-blind PET study of schizo phrenic pa tients. Biol Psy chi a try 1993;33: Miceli JJ, Gunn KP, Ru bin RH, Frackowiak RSJ, Wil liams SA, Fischman A. 5HT2 and D2 re cep tor oc cu pancy of ziprasidone in healthy vol un teers. Am Coll Neuropsychopharmacol 1996;35: Bench CJ, Lammertsma AA, Grasby PM, Dolan RJ, Warrington SJ, Boyce M, and oth ers. The time course of bind ing to striatal do pa mine D2 re cep tors by the neuroleptic ziprasidone (CP-88,059-01) de ter mined by pos i tron emis sion to mog - ra phy. Psychopharmacology 1996;124: Kasper S, Tauscher J, Küfferle B, Asenbaum S, Barnas C, Hesselmann B, and oth ers. Sertindole and do pa mine D2 re cep tor oc cu pancy. Biol Psychiatry1997;42:161S. 45. Kasper S, Tauscher J, Küfferle B, Barnas C, Hesselmann B, Asenbaum S, and oth ers. Sertindole and do pa mine D2 re cep tor oc cu pancy in com par i son to risperidone, clozapine and haloperidol. Psychopharmacology 1998;136: Kapur S, Zipursky R, Remington G, Jones C, McKay G, Houle S. PET ev i dence that loxapine is an equipotent blocker of 5-HT2 and D2 re cep tors: Im pli ca tions for the ther a peu tics of schizo phre nia. Am J Psy chi a try 1997;154: Busatto GF, Pilowsky LS, Costa DC, Ell PJ, Verhoeff NPLG, Kerwin RW. Do - pa mine D2 re cep tor block ade in vivo with the novel antipsychotics risperidone and remoxipride an 123I-IBZM sin gle pho ton emis sion to mog ra phy (SPET) study. Psychopharmacology 1995;117: Klemm E, Grünwald F, Kasper S, Menzel C, Broich K, Danos P, and oth ers. [123I]IBZM SPECT for im ag ing of striatal D2 do pa mine re cep tors in 56 schizo - phrenic pa tients tak ing var i ous neuroleptics. Am J Psy chi a try 1996;153: Scherer J, Tatsch K, Schwarz J, Oertel WH, Konjarczyk M, Albus M. D2-dopamine re cep tor oc cu pancy dif fers be tween pa tients with and with out extrapyramidal side ef fects. Acta Psychiatr Scand 1994;90: Pilowsky LS, Busatto GF, Tay lor M, Costa DC, Sharma T, Sigmundsson T, and oth ers. Do pa mine D2 re cep tor oc cu pancy in vivo by the novel atyp i cal antipsychotic olanzapine a 123I IBZM sin gle pho ton emis sion to mog ra phy (SPET) study. Psychopharmacology 1996;124: Su T-P, Breier A, Coppola R, Hadd K, Elman I, Adler C, and oth ers. D2 re cep tor oc cu pancy in risperidone and clozapine-treated schizo phren ics. Biol Psy chi a try 1996;39: Su T-P, Breier A, Coppola R, Hadd K, Elman I, Al der C, and oth ers. D2 re cep tor oc cu pancy dur ing risperidone and clozapine treat ment in chronic schizo phre nia: re la tion ship to blood level, ef fi cacy and EPS. Soc Neurosci Abstr 1996;22: Gefvert O, Lundberg T, Wieselgren I-M, Hagström P, Bergström M, Långström B, and oth ers. D2 and 5-HT2A re cep tor bind ing of dif fer ent doses of quetiapine in schizo phren ics: a PET study. Am Coll Neuropsychopharmacol 1997;36: Karbe H, Wienhard K, Hamacher K, Huber M, Herholz K, Coenen HH, and oth - ers. Pos i tron emis sion to mog ra phy with (18F) methylspiperone dem on strates D2 do pa mine re cep tor bind ing dif fer ences of clozapine and haloperidol. J Neu ral Transm 1991;86: Louwerens JW, Buddingh JA, Zijlstra S, Pruim J, Korf J, Paans AMJ, and oth ers. Do pa mine (D2)-re cep tor oc cu pancy in clozapine-treated pa tients as mea sured by pos i tron emis sion to mog ra phy us ing 18FESP. In: Brunello N, Mendlewicz J, Racagni G, ed i tors. New gen er a tion of antipsychotic drugs: novel mech a nisms of ac tion. Vol ume 4. Basel: Karger; p Louwerens JW, Slooff CJ, Korf J, Coppens HJ, Paans AMJ. Dopamine2- and serotonin2-receptor-antagonism by antipsychotics in man. Schizophr Res 1996;18: Conley R, Medoff D, Wong D, Tamminga C. 11C NMSP re cep tor oc cu pancy by clozapine and haloperidol in schizo phrenic sub jects. Schizophr Res 1995;15: Conley R, Zhao M, Wong D, Tamminga C. 11C NMSP re cep tor oc cu pancy by clozapine and haloperidol in schizo phrenic sub jects. Biol Psy chi a try 1996;39: Pickar D, Su T-P, Weinberger DR, Coppola R, Malhotra AK, Knable MB, and oth ers. In di vid ual vari a tion in D2 do pa mine re cep tor oc cu pancy in clozapine-treated pa tients. Am J Psy chi a try 1996;153: Brunello N, Masotto C, Steardo L, Markstein R, Racagni G. New in sights into the bi ol ogy of schizo phre nia through the mech a nism of ac tion of clozapine. Neuropsychopharmacology 1995;13: Kapur S, Seeman P. Antipsychotic agents dif fer in how fast they come off the do pa mine D2 re cep tors. Im pli ca tions for atyp i cal antipsychotic ac tion. J Psy chi a - try Neurosci 2000;25: Kapur S, Seeman P. Does fast dis so ci a tion from the do pa mine D2 re cep tor ex - plain the ac tion of atyp i cal antipsychotics? A new hy poth e sis. Am J Psy chi a try 2001;158: Kapur S, Zipursky R, Jones C, Shammi CS, Remington G, Seeman P. A pos i tron emis sion to mog ra phy study of quetiapine in schizo phre nia a pre lim i nary find - ing of an antipsychotic ef fect with only tran siently high do pa mine D2 re cep tor oc cu pancy. Arch Gen Psy chi a try 2000;57: Meltzer HY, Lee MA, Ranjan R, Ma son EA, Cola PA. Re lapse fol low ing clozapine with drawal: Ef fect of neuroleptic drugs and cyproheptadine. Psychopharmacology (Berl) 1996;124: Thyrum PT, Yeh C, Potkin SG, and oth ers. Safety and tolerability of switch ing from con ven tional antipsychotic ther apy to Seroquel (quetiapine fumarate) fol - lowed by abrupt with drawal from Seroquel. Am Col lege Neuropsychopharmacol 1997;36: Chouinard G, Kopala L, Labelle A, Beauclair L, John son SV, Singh KI. Phase-IV multicentre clin i cal study of risperidone in the treat ment of out pa tients with schizo phre nia. The RIS-CAN-3 study group. Can J Psy chi a try 1998;43: Seeman P. Tardive dyskinesia, do pa mine re cep tors, and neuroleptic dam age to cell mem branes. J Clin Psychopharmacol 1988;8:3S 9S. 68. Kapur S, Seeman P, Zipursky R, Remington GJ. Fast dis so ci a tion from the do pa - mine D2 re cep tor (not high af fin ity at mul ti ple re cep tors) is the key to atyp i cal antipsychotics. Schizophr Res 2001;49 (Suppl 1,2): Meltzer HY, Matsubara S, Lee J-C. Clas si fi ca tion of typ i cal and atyp i cal antipsychotic drugs on the ba sis of do pa mine D-1, D-2 and serotonin2 pki val - ues. J Pharmacol Exp Ther 1989;251: Meltzer HY, Nash JF. Ef fects of antipsychotic drugs on se ro to nin re cep tors. Pharmacol Rev 1991;43: Wadenberg M-LG, Browning JL, Young KA, Hicks PB. An tag o nism at 5-HT2A re cep tors po ten ti ates the ef fect of haloperidol in a con di tioned avoid ance re - sponse task in rats. Pharmacol Biochem Behav 2001;68: Seeman P, Corbett R, Van Tol HHM. Atyp i cal neuroleptics have low af fin ity for do pa mine D2 re cep tors or are se lec tive for D4. Neuropsychopharmacology 1997;16: and 1997;16: Seeman P, Kapur S. Olanzapine bind ing to do pa mine re cep tors in vi tro and in vivo. In: Olanzapine (Zyprexa) a novel antipsychotic. Tran PV, Bymaster F, Tye N, Herrera J, Breier A, Tollefson G, Eli Lilly and Co, ed i tors. Phil a del phia (PA): Lippincott Wil liams and Wilkins; p Nyberg S, Eriksson B, Oxenstierna G, Halldin C, Farde L. PET study of D2- and 5-HT2 re cep tor oc cu pancy in duced by risperidone in schizo phrenic pa tients. Am Coll Neuropsychopharmacol 1996;35: Nyberg S, Nakashima Y, Nordström A-L, Halldin C, Farde L. Pos i tron emis sion to mog ra phy of in-vivo bind ing char ac ter is tics of atyp i cal antipsychotic drugs. Re view of D2 and 5-HT2 re cep tor oc cu pancy stud ies and clin i cal re sponse. Br J Psy chi a try 1996;168: Trichard C, Paillere-Martinot ML, At tar-levy D, Recassens C, Monnet F, Martinot JL. Bind ing of antipsychotic drugs to cor ti cal 5-HT2A re cep tors: a PET study of chlorpromazine, clozapine, and amisulpride in schizo phrenic pa tients. Am J Psy chi a try 1998;155: Invernizzi RW, Cervo L, Samani R. 8-Hydroxy-2-(Di-N-propylamino) tetralin, a se lec tive se ro to nin-1a re cep tor ag o nist, blocks haloperidol-induced cat a lepsy by an ac tion on raphe nu clei medianus and dorsalis. Neuropharmacology 1988;27: Wadenberg M-L. An tag o nism by 8-OH-DPAT, but not ritanserin, of cat a lepsy in duced by SCH23390 in the rat. J Neu ral Transm 1992;89: Weiner DM, Burstein ES, Nash N, Croston GE, Cur ri er EA, Vanover KE, and oth ers. 5-hydroxytryptamine-2A re cep tor in verse agonists as antipsychotics. J Pharmacol Exp Ther 2001;299: Remington G, Kapur S, Shammi S, Seeman P. Ex tended dos ing as an al ter na - tive to in ter mit tent or tar geted antipsychotic ther apy: Ra tio nale and pi lot study. Schizophr Res 2001;49 (Suppl 1,2):243. W Can J Psy chia try, Vol 47, No 1, Feb ru ary