@
`
`Weight gain, antipsychotic drug treatment
`and pharmacogenomics
`
`
`
` Q'
`Til/l
`ll"mfill-Tlif
`i
`ll
`
`Gavin P Reflalds
`Department afBiomedica]
`Science, Universityaf
`Sheffield, ”firm Bank,
`Shefi'icld, 510 ZTN, UK
`Tel: +44 (0)114 222 4662;
`Fax.‘ +44 (0)114 276‘ .5413;
`E—mail.’ GBRemaIdrCfl’
`rhefiieldacmlr
`
`Ashley Publications Ltd
`wwwashley—pub.com
`
`‘It would be surprising and
`
`disappointing, if in a few years'
`
`time genetic testing for weight gain
`
`susceptibility was not a routine
`
`procedure prior to drug treatment
`
`of psychiatric patients.’
`
`Pharmacotherapy is almost inevitably associated
`with unwanted side effects. This is particularly
`true for psychiatric drugs, which are notably
`associated with a wide variety of common, often
`severe and limiting, side effects including seda-
`tion,
`lethargy, hypotension, autonomic effects,
`both acute and chronic motor symptoms and
`weight gain.
`The tolerability of such side effects relates to a
`variety of factors, which include social attitudes
`as well as perceptions of risk and benefit. There
`are often substantial individual and occasionally
`ethnic differences in the susceptibility to side
`effects, pointing to pharmacogenomic influ-
`ences. Minimizing the incidence of such side
`effects has been a major target of drug develop-
`ment, although as yet little attention has been
`paid to the potential that pharmacogenomics
`may offer to understanding treatment—induced
`side effects.
`
`the extrapyramidal side effects
`Classically,
`(EPS) have been considered to be the most
`troublesome of those induced by antipsychotics
`and include:
`
`0 parkinsonism
`' akathisia (motor restlessness)
`' dyskinesias
`
`As freedom from EPS has been a major target in
`the development of new drugs, there are now
`several antipsychotic treatments available that
`have a lower incidence of these side effects.
`
`These ‘atypical’ drugs have proven invaluable for
`many patients, although they have yet to entirely
`replace cheaper classical antipsychotic drugs.
`However, with increasing use of newer atypical
`antipsychotics and diminishing incidence of
`EPS, other problematic side effects have become
`more apparent. One of the most disabling of
`these, weight gain, will not only influence toler-
`
`ance of and hence compliance with, drug treat—
`ment but is inevitably associated with substantial
`morbidity. This includes diabetes, hypertension
`and cardiovascular disease, along with other con-
`sequences of obesity, such as joint disease. Two
`atypical antipsychotic drugs, clozapine and olan-
`zapine, may induce particularly profound weight
`gain although few of the other antipsychotics are
`free of this effect
`[1]. However, drug induced
`weight gain is not unique to the antipsychotics. A
`variety of other drugs can induce this side effect
`including some antidepressants and mood stabi-
`lizers and the antimigraine 5-HT1D agonists.
`The underlying mechanisms are far from
`understood. The pharmacological profile of the
`antipsychotic drugs has provided clues, with
`effects at a variety of neurotransmitter receptors
`being implicated [2]. 5-HT systems have long
`been known to be associated with feeding and
`appetite, primarily via influences at the hypotha-
`lamus. Almost all the newer atypical antipsy-
`chotics have high affinities for the 5-HT2A
`receptor and several also affect
`the 5-HT2C
`receptor site. This latter receptor has been par—
`ticularly implicated in weight gain since the
`finding that gene knockout of the 5-HT2C
`receptor
`in mice can result
`in obesity and
`increased feeding [3]. Clozapine and olanzapine
`are high-affinity 5-HT2C antagonists likely to
`block this receptor site at normal clinical doses,
`and this action could certainly contribute to
`their propensity to induce weight gain.
`The strong, if circumstantial, evidence for the
`involvement of 5-HT in feeding behavior has
`provided a valuable source for candidate gene
`hypotheses. The importance of 5-HT systems in
`the treatment of psychiatric disorders has stimu-
`lated the investigation of 5-HT—related candidate
`genes in other aspects of neuropsychiatry. The
`polymorphisms investigated in these candidate
`genes are often functional. For example,
`the
`common insertion/deletion polymorphism of
`the promoter region of the 5-HT transporter
`gene has been associated with affective disorder
`[4], while the structural coding region and pro-
`moter polymorphisms of the 5-HT2A receptor
`gene are associated with schizophrenia and
`antipsychotic drug response [5].
`
`2002 © Ashley Publications Ltd ISSN 1462-2416
`
`Pharmacogenomics (2002) 3(5), 567-570
`
`1
`
`567
`Exhibit 2029
`
`Slayback v. Sumitomo
`lPR2020-01053
`
`1
`
`Exhibit 2029
`Slayback v. Sumitomo
`IPR2020-01053
`
`

`

`EDITORIAL
`
`568
`
`There have been some recent attempts to
`apply such pharmacogenetic hypotheses
`to
`understanding individual variability in drug-
`induced weight gain. A study of several polymor-
`phisms of genes relating to 5-HT neurotransmis—
`sion, including those of the 5-HT transporter
`and several 5-HT receptors failed to identify any
`association with weight gain induced by clozap-
`ine [6]. Another investigation of nine candidate
`genes,
`including three 5-HT receptors, also
`failed to identify any association with clozapine-
`induced weight gain [7].
`
`‘Some preliminary studies using
`
`reporter gene assays suggested
`
`that the likely mechanisms relate to
`
`differences in promoter activity,
`
`resulting in differences in levels of
`receptor expression.’
`
`Previous genetic studies of the 5-I‘IT2C recep—
`tor in schizophrenia and its treatment have con-
`centrated primarily on the Cys23Ser (68C/G)
`polymorphism. This has also been implicated as
`a genetic factor in antipsychotic drug response
`[8], although it is not associated with abnormal
`body weight [9], nor is there a significant associa-
`tion with clozapine—induced weight gain [7]. Its
`very low frequency in a Chinese population cer—
`tainly rules it out as important in the genetic
`control of drug-induced weight gain in this eth-
`nic group. However, a recent pharmacogenetic
`study [10] identified several haplotypes of the pro-
`moter region of the 5'HTZC receptor gene; these
`comprised four polymorphisms:
`a variable
`length tandem repeat (VLTR) in linkage disequi-
`librium with three SNPs.
`Interestingly,
`these
`were found to be associated with obesity and dia-
`betes in a Japanese population [10]. Some prelim-
`inary studies using reporter gene assays suggested
`that the likely mechanisms relate to differences
`in promoter activity, resulting in differences in
`levels of receptor expression. This was an excit—
`ing observation in that it provided strong evi-
`dence for the 5'HT2C receptor as a candidate for
`the genetic control of drug-induced weight gain.
`The results from a study of one of these SNPs
`(—759C/T) in a series of 123 Chinese first-epi-
`sode patients with schizophrenia have recently
`been published [11]. It was found that this genetic
`marker differentiated effects on body weight fol-
`lowing antipsychotic drug treatment, whereby
`presence of the variant —759T allele appeared to
`
`be protective. This was best illustrated by using
`the standard Food and Drug Administration
`(FDA) criterion of a 7% increase as indicative of
`a clinically significant weight gain. After 6 weeks
`of treatment with antipsychotic drugs, 28% of
`the 96 patients with the —759C allele, but none
`of those carrying the T allele, met this criterion;
`after 10 weeks of treatment the corresponding
`proportions were 51% and 15%, yielding an
`odds ratio of 6.0. The effect was apparent
`whether the patients received either chlorpro-
`mazine, a classical antipsychotic, or the atypical
`risperidone. A further analysis (unpublished) of a
`subgroup of these first-episode patients switched
`to clozapine, one of the two worst offenders in
`inducing weight gain, showed that with this
`drug too the —759 SNP was strongly associated
`with weight gain.
`These data yielded some further interesting
`observations relating to the X—linkage of the
`5-HT2C receptor gene. While homozygous C/C
`females and hemizygous C allele males both
`demonstrated substantial weight gain,
`female
`(C/T)
`patients
`carrying the variant
`allele
`appeared to show a greater weight gain than their
`male (hemizygous T) counterparts. This is likely
`to reflect the contributions from both the C and
`
`T alleles to the female phenotype.
`These findings clearly implicate neurotrans-
`mitter action through the 5'HTZC receptor in
`the pathophysiological mechanisms underlying
`weight gain following initial antipsychotic drug
`treatment. However, what exactly those mecha-
`nisms might be remains elusive. Although the
`drugs that are most problematic demonstrate
`substantial 5'HTZC antagonism, some with rela-
`tively weak effects at this receptor still induce
`weight gain. An example of such a drug is risp-
`eridone, where the weight gain is nevertheless
`still apparently under the control of the 5-HT2C
`promoter polymorphism.
`Thus, the pharmacology underlying antipsy-
`chotic drug-induced weight gain may involve a
`variety of effects, not necessarily always includ-
`ing 5—I-IT2C antagonism, and these pharmaco—
`logical mechanisms can be differentiated from
`the pharmacogenomic control provided by the
`5-HT2C promoter polymorphism. How exactly
`might this control be exerted? The fact that the
`promoter SNPs contribute to haplotypes with
`functional differences provides a strong indica-
`tion of what might be the mediating ‘endophe-
`notype’. The variant
`form reportedly has a
`higher promoter activity, presumably leading to
`higher
`receptor expression [10], although it
`
`Pharmacogenomics (2002) 3(5)
`
`2
`
`

`

`should be mentioned that a recent report failed
`to find an effect of the VLTR polymorphism on
`promoter activity [12]. Whether or not the geno-
`typic differences related to differences in receptor
`density (which would need to be established in
`brain tissue) and hence to differences in 5-HT2C-
`mediated neurotransmission, it is still hard to
`
`explain differential effects on weight gain in
`patients receiving clozapine, which would fully
`occupy 5-HT2C receptors. Thus, more complex
`effects than genotypes producing differences in
`numbers of receptors need to be invoked. One
`explanation might be that the 5-HT2C receptor-
`mediated control of weight gain is relatively less
`important in subjects carrying the —759T allele,
`presumably as a result of development differ-
`ences consequent upon the immediate effects of
`the promoter polymorphisms. This certainly
`provides
`some testable working hypotheses
`relating to, for example, measures of hypotha-
`lamic 5'HTZC function.
`Nevertheless, whatever the underlying mecha-
`nism of the differences in weight gain between
`genotyes, the observation provides a potential
`test for the propensity to weight gain in patients
`being prescribed antipsychotic drugs. It is likely,
`although yet to be shown, that the finding will
`generalize to the weight gain observed with a
`range of other drug treatments. These include:
`
`'
`
`'
`
`the antidepressant mirtazepine, which is a
`
`5-HT2C antagonist
`the serotonin re-uptake inhibitors
`
`' other drugs with effects on 5-HT systems,
`such as lithium and the 5-HT“) agonists
`
`Of course there are many influences on drug-
`induced weight gain, and the 5-HT2C receptor
`gene polymorphism is only one of a variety of
`genetic factors that may have effects on appetite,
`satiety,
`fat metabolism and disposition,
`thirst,
`activity and so on. The behavioral mechanism
`behind weight gain induced by antipsychotic
`drugs is not understood, although evidence points
`to a supression of satiety, at least with olanzapine
`[13]. This suggests in turn an effect on hypotha-
`lamic function, possibly through the action of
`5'HTZC receptors on the satiety response.
`
`EDITORIAL
`
`
`
`‘Weight gain is not solely under
`
`genetic control; environmental
`
`factors inevitably interact with
`
`genomic influences to modify their
`effects on the organism.’
`
`Although the association with the —759C/T
`polymorphism is a strong one, it only explains a
`proportion of the variance in drug-induced
`weight gain. Functional polymorphisms
`in
`other neurotransmitter receptors that mediate
`hypothalamic function may also play important
`roles. But weight gain is not solely under genetic
`control; environmental factors inevitably inter—
`act with genomic influences to modify their
`effects on the organism. This environmental
`interaction with genetics is no better illustrated
`than by the current obesity epidemic that,
`despite the strong genetic component, can be
`understood in terms of a variety of recent and
`rapid social and economic changes.
`Drug-induced weight gain will also be influ-
`enced by non-genetic factors. The drug treat-
`ment
`itself can be modified;
`for example,
`patients can be prescribed antipsychotic drugs
`with fewer or minimal effects on weight. Dietary
`restriction and education can also be valuable.
`
`identifying
`that
`there is no doubt
`However,
`potential genetic susceptibility to weight gain in
`patients would provide an invaluable addition to
`the clinical information from which pharmaco-
`therapeutic decisions can be made.
`It is unfortunate that genetic testing is not yet
`an accepted and routine part of the process in
`determining optimal and individualized pharma-
`cotherapy, despite the clear potential that phar-
`macogenomics has
`to offer. CYP2D6 and
`apolipoprotein E status are just two of many
`other examples of genetic measures that could
`provide valuable information for drug prescrib-
`ing. It would be surprising and disappointing, if
`in a few years’ time genetic testing for weight gain
`susceptibility was not a routine procedure prior
`to drug treatment of psychiatric patients. Then
`pharmacogenomics will truly have come of age.
`
`www. pharmaco-genomics.co. uk
`
`569
`
`3
`
`

`

`EDITORIAL
`
`Bibliography
`Papers of special note have been highlighted as
`either of interest (°) or of considerable interest (")
`to readers.
`1.
`
`Allison DB, Mentore JL, Heo M at 511.:
`Antipsychotic-induced weight gain: a
`comprehensive research synthesis. Am. ]
`Hychia. 156, 1686—1696 (1999).
`A comprehensive meta-analysis of the
`different propensities of the antipsychotic
`drugs to induce weight gain.
`Casey DE, Zorn SH: The pharmacology of
`weight gain with antipsychotics. _[ CII'II.
`Hyrlzz‘a. 62(Suppl. 7), 4-10 (2001).
`Tecott LH, Sun LM, Akana SF etal: Eating
`disorder and epilepsy in mice lacking 5—
`HTZC serotonin receptors. Nature 374,
`542-546 (1995).
`An important receptor knockout model
`highlighting the role of the 5—HT2C
`receptor in the control of body weight.
`Lesch KP: Variation of serotonergic gene
`expression: neurodevelopment and the
`complexity of response to
`psychopharrnacologic drugs. Eur.
`Mmprychoplratmaml. 11, 457-474 (2001).
`A useful review of the different genomic
`approaches to investigating the 5-HT
`system and its pharmacogenetics, including
`a discussion of 5-HT pharmacogenomics
`in a developmental context.
`Arranz M], Munro], Owen M] eta];
`Evidence for association between
`
`polymorphisms in the promoter and coding
`
`regions of the 5—HT2A receptor gene and
`response to clozapine. A101. ngrhia. 3, 61-66
`(1998).
`Hong C], Lin CH, Yu YWY, Yang KH,
`Tsai SJ: Genetic variants of the serotonin
`system and weight change during clozapine
`treatment. Plramawgenetim 11, 265-268
`(200 1) .
`Basile VS, Masellis M, McIntyre RS,
`Meltzer HY, Lieberman IA, Kennedy JL:
`Genetic dissection of atypical antipsychotic-
`induced weight gain; novel preliminary data
`on the pharmacogenetic puzzle. ,1 Clin.
`Rychz‘a. 62(Suppl. 23), 46-66 (2001).
`A useful and comprehensive review of the
`complexities of hypothalarnic control of
`body weight, including (negative)
`association data testing some candidate
`gene hypotheses of weight gain induced
`by clozapine.
`Sodhi MS, Arranz M], Curtis D et a1:
`Association between clozapine response and
`allelic variation in the 5-HT2C receptor
`gene. Mumwporfl, 169—172 (1995).
`An early observation of a genetic factor
`contributing to antipsychotic drug response.
`Lentes KU, Hinney A, Ziegler A eta];
`Evaluation of a CysZSSer mutation within
`the human 5—HT2C receptor gene: no
`evidence for an association of the mutant
`
`allele with obesity or underweight in
`children, adolescents and young adults. Life
`Sci. 61, PL9—P16 (1997).
`Yuan X, Yarnada K, Ishiyama—Shigemoto S,
`
`10.
`
`Koyarna W, Nonaka K: Identification of
`polymorphic loci in the promoter region of
`the serotonin 5HT2C receptor gene and
`their association with obesity and Type II
`diabetes. Diabetolagia 43, 373—376 (2000).
`First identification of the importance of
`these functional polymorphisms in control
`of body weight.
`Reynolds GP, Zhang Z], Zhang XB:
`Association of antipsychotic drug-induced
`weight gain with a 5-HT2C receptor gene
`polymorphism. Lancet359, 2086—2087
`(2002).
`Demonstration of a strong association of a
`polymorphism of the promoter region of
`the 5-HT2C receptor with antipsychotic
`drug-induced weight gain in previously
`untreated schizophrenic patients.
`Meyer J, Saam W, Mossner R eta];
`Evolutionary conserved rnicrosatellites in
`the promoter region of the 5-
`hydroxytryptamine receptor 2C gene
`(HTRZC) are not associated with bipolar
`disorder in females. 1 Mural Fatwa. 109,
`939—946 (2002).
`Thomton-Jones Z, Neill JC, Reynolds CF:
`The atypical antipsychotic olanzapine
`enhances ingestive behaviour in the rat. I
`Psychopbarmaca]. 16, 35—37 (2002).
`A demonstration of the acute effect of
`olanzapine in increasing feeding via a
`supression of satiety.
`
`11.
`
`12.
`
`13.
`
`570
`
`Pharmacogenomics (2002) 3(5)
`
`4
`
`