A feasibility and tolerability study of lithium in Alzheimer's disease

A feasibility and tolerability study of lithium in Alzheimer's disease
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A feasibility and tolerability study of lithiumin Alzheimer’s disease Alastair Macdonald, Kate Briggs, Michaela Poppe, Andrea Higgins,Latha Velayudhan and Simon Lovestone*  NIHR Biomedical Research Centre for Mental Health at the South London and Maudsley Foundation NHS Trust and the MRC Centre for Neurodegeneration Research, Institute of Psychiatry, King’s College London, London, UK  SUMMARY Objective  To assess the safety and feasibility of prescribing long term lithium to elderly people with mild to moderateAlzheimer’s disease (AD). Methods  An open label treatment group with low dose lithium for up to 1 year with the Lithium Side Effects Rating Scaleas the primary outcome measure. A comparison group matched for cognition and age not receiving lithium therapy. Results  Twenty-twopeoplewithADinitiatedlithium.Fourteenparticipantsdiscontinuedtherapyafterameanof16weeksof treatment compared to the 39 weeks for those continuing to take treatment at the end of the study. Three patientsdiscontinued treatment due to possible side effects that abated on ceasing therapy. The reports of side effects on the primaryoutcome scale did not differ between those discontinuing therapy and those remaining in the study. Two patients died whilstreceiving lithium––in neither case was the treatment felt to be related to cause of death. There was no difference in deaths,drop outs or change in MMSE between those receiving lithium and the comparison group. Conclusions  Lithium treatment in elderly people with AD has relatively few side effects and those that were apparentlydue to treatment were mild and reversible. Nonetheless discontinuation rates are high. The use of lithium as a potentialdisease modification therapy in AD should be explored further but is not without problems. Copyright # 2008 John Wiley&Sons, Ltd. key words —Alzheimer’s disease; lithium; trial; GSK-3; disease modifying INTRODUCTIONAlzheimer’s disease (AD) is the commonest of the dementias that together affect over 24 millionpeople worldwide (Ferri  et al ., 2005). Although therearecurrentlynodiseasemodifyingtherapies,advancesin understanding of the molecular pathogenesis haveidentified two key therapeutic targets. The first of these is amyloid, the peptide that aggregates in theneuritic plaque and a variety of compounds designedto modify the generation, aggregation or clearance of amyloid are in development. An alternative potentialtherapeutic target in AD is the neurofibrillary tangle,formed from aggregated and phosphorylated tauprotein. One approach to this therapeutic target wouldbe to reduce the phosphorylation of tau throughinhibition of the relevant kinase.Whilst many kinases can phosphorylate tau invitro,increasing evidence suggests that glycogen synthasekinase-3 (GSK-3) is the predominant tau-kinase inbrain [reviewed in (Mudher and Lovestone, 2002;Bhat  et al ., 2004a)] and GSK-3 levels and activity arealtered in people with AD (Hye  et al ., 2005) and inneurons affected by AD pathology (Pei  et al ., 1999).Moreover, over-expression of GSK-3 in mice inducesneurodegeneration (Lucas  et al ., 2001; Hernandez et al ., 2002) and over-expression of GSK-3 inDrosophila induces aggregation of tau into tanglessimilar to those of AD (Jackson  et al ., 2002). Inhi-bition of GSK-3 activity restores neuronal function inboth mouse (Noble  et al ., 2005; Hooper  et al ., 2007; INTERNATIONAL JOURNAL OF GERIATRIC PSYCHIATRY  Int J Geriatr Psychiatry  2008;  23 : 704–711.Published online 8 January 2008 in Wiley InterScience(  DOI : 10.1002/gps.1964*Correspondence to: S. Lovestone, Department of PsychologicalMedicine, King’s College London, Institute of Psychiatry, DeCrespigny Park, London, SE5 8AF, UK.E-mail: Copyright # 2008 John Wiley & Sons, Ltd.  Received 19 September 2007  Accepted 6 November 2007   Rockenstein  et al ., 2007) and fly (Mudher et al ., 2004)models of neurodegeneration making GSK-3 inhi-bition a key goal in AD therapeutic research (Huangand Klein, 2006; Mazanetz and Fischer, 2007).Although specific and potent GSK-3 inhibitorshave been developed, one drug commonly used inpsychiatry, lithium, is also a GSK-3 inhibitor (Jope,2003). The therapeutic range for lithium in man is0–5–1.5mM and lithium has a Ki for GSK-3 of 2mMsuggesting that partial GSK-3 inhibition is likely to beachieved during therapy. Effects of lithium on tau canbe measured in a dose-response manner to levels aslow as 0.1mM (Leroy  et al ., 2000) in vitro and withinthe normal therapeutic range (0.6mM) in mousemodels of tau-induced neurodegeneration (Noble et al ., 2005). Indeed lithium might prevent the pro-gression from amyloid-related pathology to neurode-generation as pre-treatment of both rats and rabbitswith lithium prevented neurotoxicity associated withintra-cerebral amyloid injection (De Ferrari  et al .,2003; Ghribi  et al ., 2003).These data have prompted the suggestion thatlithiummightbeapotentialdiseasemodifyingtherapyfor AD (Bhat  etal .,2004b;Aghdam and Barger, 2007).However, if lithium modifies disease progression inAD then it might be expected that long-term lithiumusers would be at lower risk of AD. Three studies haveexamined this, with confusing results. Terao  et al .(2006) found that psychiatric care patients who hadever taken lithium had a higher MMSE score thanthose who had never taken lithium. In line with thisfinding, Nunes  et al . (2007) found lower rates of dementia in patients with bipolar disorder treated withlithium than in those treated with other mood-stabilisers. However, Dunn  et al . (2005) found a slightincreaseinriskofdementia inthosetakinglithium inalarge primary care database. Confounding issuesconstraining such analyses, and perhaps contributingto contrasting results, include the fact that depressionis both a risk factor for AD and an indication forlithium therapy, and that lithium therapy is oftendiscontinued in late life. It is difficult to see how theseobstacles could be overcome in an observational studyof any size.In summary the evidence that GSK-3 phosphoryl-ation of tau precedes and promotes tangle formationand loss of neuronal function is strong and theevidence that lithium inhibits GSK-3 function attherapeutic levels overwhelming. Together these datasuggest lithium as a therapeutic strategy. Diseasemodification trials in AD are typically more than12 months in duration and chronic treatment for anysuccessful drugs inevitable. However, lithium has anarrow therapeutic window and known neurotoxic andother side effects, especially in the elderly. Further-more, an extended period of dose-finding andstabilisation is necessary and even when stabilised,patients require regular monitoring of serum lithiumlevels. Interactions with many other drugs oftenprescribed for the elderly have been reported. All of these factors might limit the usefulness of lithium inAD, especially in the frail and elderly and in thosewith other co-existing illness. We therefore conducteda pilot study assessing the feasibility and tolerabilityof lithium carbonate at therapeutic levels in mild tomoderate AD over an extended period. This pilotstudy was an open label, pragmatic and observationalstudy of a series of patients treated for up to 1 yearwith low dose lithium. Our aims were to assess theability of patients with mild to moderate AD totolerate lithium therapy and to assess the frequency of adverse effects.METHOD Protocol The trial was an open label pragmatic trial of up to1 year for a maximum of 25 subjects with mildto moderate AD. Three phases of the trial wereplanned––screening, stabilisation and treatment. Allsubjects thought by Mental Health of Older Adultsclinical teams in the South London and MaudsleyNHS Trust to have mild to moderate dementiawere eligible for screening. In total 480 subjectswere screened for inclusion by examination of casenotes and preliminary discussion with carers. Thoseincluded in the trial were assessed at baseline with aclinical examination together with scales to assesscognition [Mini Mental State examination; MMSE;Folstein  et al . (1975) and Alzheimer’s Disease Asses-sment Scale––cognitive section; ADAS-cog; Rosen et al . (1984)], function [Bayer Activities of DailyLiving Scale; Hindmarch  et al . (1998)], behaviour[Neuropsychiatric Inventory; NPI; Cummings  et al .(1994)] and global deterioration [Global DeteriorationScale; GDS; Reisberg  et al . (1982)]. In addition, allpatients had an ECG and an adapted Lithium SideEffects Rating Scale (LISER; Haddad  et al . (1999)].TheLISERisdesignedasaself-ratingscaleandinthisinstance the questions were addressed by theresearcher to the informant.Patients meeting inclusion criteria were commen-ced on low dose lithium carbonate (100mg) and werethen assessed fortnightly with lithium dose adjustedaccording to serum levels aiming to reach a steady Copyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 704–711.DOI: 10.1002/gps lithium and alzheimer’s disease  705  state of between 0.3 and 0.8mM. Once stabilisedassessments were made monthly according to theprotocolshowninTable1.Inallcasesafinalassessmentwas attempted 1 year after inclusion into the trial.The primary outcome measure was the LithiumSide Effects Scale and any other reported adverseevents or drop-out from the trial. Secondary outcomesincluded change in cognition (MMSE) or function(Bayer). Subjects We recruited patients with a diagnosis of probable orpossible NINCDS-ADRDA AD, of mild to moderateseverity (MMSE range 12–24) from a large mentalhealth NHS Trust and a memory clinic in a local acutegeneral hospital. Exclusion criteria were age less than60 years, evidence of another neurodegenerativedisorder or physical illness that would explain thecognitive impairment, contraindications to lithiumtherapy (e.g. significant renal impairment or thyroiddisease), recent stroke and lack of frequent carercontact. Those treated with anti-dementia medicationwere not excluded, nor were residents in care homes.Patients with capacity gave consent following adiscussion of the trial after which they were givenan information sheet and a period of reflection.Capacity was considered to be present if the potentialparticipant could understand the nature of the trial, usethis information to consider whether they wished toparticipate, retain the information for a period andthen discuss with the assessor the nature of the trial,including the pros and cons of participating. In thosesubjects where capacity was considered to be impai-red, and also in those who were able to consent, assentwas also sought from their closest relative.Ethical approval was granted by the Joint SouthLondon and Maudsley and the Institute of PsychiatryNHS Research Ethics Committee. The proceduresfollowedwereinaccordancewiththeethicalstandardsof the responsible committee on human experimen-tation (institutional or regional) and with the HelsinkiDeclaration of 1975, as revised in 1983. Comparison group Inordertoassessthesecondary,cognitive,outcomealltrial participants were matched to patients with ADenrolled in an ongoing study of biomarkers at the IoP.This study, previously reported (see Hye  et al ., 2006;Ellul  et al ., 2007), includes approximately 300 peoplewith AD,all ofwhom are assessed usingaverysimilarclinical assessment process and all of whom arerecruited through an identical process from the sameNHS Trust. Matching was performed blind to trialoutcome on a consecutive basis from the cohortregister selecting comparison subjects similar to trialsubjects in age ( þ  /    5 years) and baseline MMSE( þ  /   2 points). In all other respects the inclusionand exclusion criteria were the same as for the trialsubjects––those treated with anti-dementia medi-cation were not excluded, nor were residents in carehomes. For each trial participant two comparisonsubjects were identified. Progression of MMSE after1 year and continuation in the cohort study wasrecorded. Table 1. Plan of investigationsMonthAssessment Baseline Weekly tostabilisation1 2 3 4 5 6 7 8 9 10 11 End of trialClinical assessment x x x x x x x x x x x x x xECG xNINCDS-ADRDA diagnosis xHachinski scale xMMSE x x x xADAS-cog x x x xCERAD battery x x xCANTAB-PAL x x x xNPI x x xGDS x x x x xBayer x x xLithium Side effects scale x x x x x x x x x x x x xLithium levels x x x x x x x x x x x x xRenal/thyroid function x x x x x x x x x x x x x xCopyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 704–711.DOI: 10.1002/gps 706  a. macdonald  ET AL.  RESULTS  Recruitment  We initially assessed the proportion of patients in atypical Old Age Psychiatry service likelyto be enteredinto a lithium trial. During this process we screenedover 450 patients with dementia referred to the trialfrom the service. Over 35% failed to meet the entrycriteria––either because the diagnosis of AD wasquestionable or because the MMSE was not in theentry range. A large proportion (17%) declined to takepart in such a trial, often because of the frequency of assessments and the need for regular venepuncture,especially during the stabilisation phase, and in afurther 10% were considered after assessment to beunlikely to comply with the trial procedures. Asignificant proportion (13%) had either a concurrentillness or therapy that contraindicated treatment withlithium. We aimed to recruit 25 patients with AD forlithium therapy.Between December 2004 and March 2006 41patients were identified at the screening interview asmeeting trial entry criteria and 22 patients werecommenced on therapy. Of those not actually enteringthe trial nine (47%) were found to have contra-indicated therapies or concurrent illnesses betweenscreening and trial commencement or there wasevidence suggesting non-compliance (such as withprescribed medication). The CONSORT diagram isshown in Figure 1. Of the 22 patients enteringthe trial,eight patients either completed the full year of therapyor were still receiving therapy at the end of the study(July 2006). The characteristics of the sample at thestages of attrition are presented in Table 2. Side effects and withdrawal The mean time on trial for those not completing thetrial was 16.7 weeks (SD 13.8 range 1day–38 weeks)and for those completing the trial 39 weeks (SD 14.3,range 21–55 weeks). Overall the average length of treatment was 25 weeks (SD 18.4) and the trialrepresents a total of 549 patient weeks of therapy.The primary outcomes for the trial were adverseevents as reported by patients and carers and asrecorded by the Lithium Side Effects Scale. Twopatients died whilst receiving lithium. In one, thecause of death was recorded as ‘Respiratory failure,pulmonary arterial thromboemboli, chronic obstruc-tive airways disease and pulmonary oedema’ and the Figure 1. Consort diagram.Copyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 704–711.DOI: 10.1002/gps lithium and alzheimer’s disease  707  other ‘Cerebrovascular accident and pneumonia’.In neither case was it considered likely that lithiumtreatment played a part in the death.Of the remaining 12 patients who were withdrawnfrom the trial this was due to probable side effects intwo patients, whose adverse symptoms abated whenlithium was stopped (including one whose lithiumlevels were below the therapeutic range). Adverseeffects included tremor and increased confusion con-firmed on cognitive testing. One patient was with-drawn from therapy after being admitted to hospitalwith a fall. On admission she had a lithium level of 0.99mM. Relatives requested removal of treatment intwo cases, and patients in a further two, although inone case the reason was unclear and in others thereported symptoms were unlikely to be related tolithium treatment. Reasons included a possible incre-ased confusion that was not evidenced on cognitivetesting, a concern about a change in skin colour and asensation of ‘wobbly legs’. In two cases staff in thehospitalorcarehomewereunabletoguaranteeregularand accurate administration of lithium. In threepatients the investigator withdrew the patient fromthe study. In one of these, the GP started a thiazidediuretic and did not wish to use an alternative, anotherpatient became acutely ill with Clostridium difficilegastroenteritis, and in another it became impossible toobtain blood samples for lithium levels.We analysed the results of the lithium side effectsscale in three ways––by the mean for each participantacross all assessment points in the study, by thescore at the final assessment point and by the highestscore on the scale reached by each participant at anytime point. In those commencing lithium the meanlithium side effects scale score across the duration of the trial was 57.3 (SD 9.2) in those who withdrew ordied and 55.8 (SD 4.3) in those who completed thetrial. The mean final score on the scale was 56.7 (S.D.9.3) in those who withdrew and 58.0 (SD 9.8) in thosecompleted (Table 2). The mean highest score on thelithium side effect scale achieved during the trial was58.0(SD9.3)inthosewhowithdrewand59.6(SD6.8)in those who completed. There was no significantdifference in this scale however aggregated betweenthose who completed and those who did not.The secondary outcome measures were functionaland cognitive measures. There was no correlation of either the MMSE or the Bayer functional scale witheither mean lithium level achieved or with weeks ontherapy. In order to explore further the change incognition over the course of the study, subjectsreceiving lithium were compared to a comparisongroup (1:2 matching) identified and assessed using a     T   a    b    l   e    2 .    C    h   a   r   a   c   t   e   r    i   s   t    i   c   s   o    f   p   a   t    i   e   n   t   s   w    h   o   w   e   r   e   r   e    f   e   r   r   e    d ,   a   s   s   e   n   t   e    d    /   c   o   n   s   e   n   t   e    d ,   c   o   m   m   e   n   c   e    d    l    i   t    h    i   u   m ,    d    i   s   c   o   n   t    i   n   u   e    d   o   r   c   o   m   p    l   e   t   e    d   t   r    i   a    l    R   e    f   e   r   r   e    d    A   s   s   e   n   t   e    d    /   c   o   n   s   e   n   t   e    d    C   o   m   m   e   n   c   e    d    L    i   t    h    i   u   m    C   e   a   s   e    d    L    i   t    h    i   u   m   o   r    d    i   e    d    d   u   r    i   n   g   s   t   u    d   y    S   t    i    l    l   t   a    k    i   n   g    L    i   t    h    i   u   m   a   t    1   y   e   a   r   o   r   t   e   r   m    i   n   a   t    i   o   n   o    f   t   r    i   a    l      N     4    8    0    4    1    2    2    1    4    8    M   e   a   n   a   g   e    [    S    D    ]    8    1 .    1    [    7 .    3    ]    8    1 .    0    [    7 .    4    ]    8    0 .    9    [    7 .    9    ]    8    3 .    4    [    6 .    7    ]    7    6 .    6    [    8 .    3    ]    N   o    f   e   m   a    l   e    (    %   o    f   g   r   o   u   p    )    3    0    7    (    6    4    %    )    2    8    (    6    9    %    )    1    3    (    5    9    %    )    1    0    (    7    1    %    )    3    (    3    8    %    )    M   e   a   n   y   e   a   r   s   o    f   e    d   u   c   a   t    i   o   n    [    S    D    ]  —    1    0 .    7    [    4 .    3    ]    1    1 .    4    [    5 .    3    ]    1    1 .    8    [    6 .    4    ]    1    0 .    8    [    3 .    8    ]    B   a   s   e    l    i   n   e   :    M   e   a   n    H   a   c    h    i   n   s    k    i    S   c   o   r   e    [    S    D    ]  —  —    1 .    6    [    1 .    2    ]    2    [    1 .    1    ]    0 .    8    [    0 .    9    ]    M   e   a   n    M    M    S    E   s   c   o   r   e    [    S    D    ]  —    1    7 .    4    [    5 .    4    ]    1    7 .    1    [    3 .    5    ]    1    6 .    4    [    2 .    8    ]    1    8 .    5    [    4 .    3    ]   m   e   a   n    A    D    A    S  -    C    O    G   s   c   o   r   e    [    S    D    ]  —  —    2    9 .    5    [    1    0 .    8    ]    3    3 .    9    [    1    1 .    5    ]    2    5 .    1    [    8 .    5    ]    T   o   t   a    l    C    E    R    A    D    S   c   o   r   e    [    S    D    ]  —  —    3    0 .    3    [    1    1 .    6    ]    2    4 .    2    [    1    1 .    8    ]    3    7 .    8    [    6 .    1    ]   m   e   a   n    N    P    I   s   c   o   r   e    [    S    D    ]  —  —    1    0 .    2    [    1    6 .    0    ]    1    1 .    5    [    1    9 .    2    ]    8 .    3    [    1    0 .    4    ]   m   e   a   n    G    D    S   s   c   o   r   e    [    S    D    ]  —  —    2 .    5    [    1 .    6    ]    2 .    5    [    1 .    8    ]    2 .    4    [    1 .    5    ]    M   e   a   n    B   a   y   e   r    A    D    L    S   c   o   r   e    [    S    D    ]  —  —    7 .    3    [    1 .    8    ]    8 .    0    [    1 .    3    ]    6 .    2    [    2 .    0    ]    L    i   t    h    i   u   m    S    i    d   e    E    f    f   e   c   t   s    S   c   a    l   e    (    fi   n   a    l   a   s   s   e   s   s   m   e   n   t    )  —  —    5    5 .    1    [    7 .    6    ]    5    6 .    7    [    9 .    3    ]    5    8 .    0    [    9 .    8    ]    N   u   m    b   e   r   a   c    h    i   e   v    i   n   g   t    h   e   r   a   p   e   u   t    i   c    l   e   v   e    l    (   m   e   a   n    d   a   y   s    f   r   o   m   c   o   m   m   e   n   c   e   m   e   n   t    [    S    D    ]    )  —  —    1    4    (    3    5 .    5    [    2    2 .    0    ]    )    6    (    2    3 .    7    [    1    3 .    3    ]    )    8    (    4    4 .    5    [    2    3 .    7    ]    )    M   e   a   n   t    i   m   e   o   n    l    i   t    h    i   u   m    (   w   e   e    k   s    )    [    S    D   ;   r   a   n   g   e    ]  —  —    2    5    [    1    8 .    0    4   ;    0 .    1  –    5    5    ]    1    6 .    7    [    1    3 .    8   ;    0 .    1  –    3    9    ]    3    9 .    4    [    1    4 .    3   ;    2    1  –    5    5    ]    M   e   a   n    l    i   t    h    i   u   m    l   e   v   e    l    (    S    D   ;    R   a   n   g   e    )    0 .    4    0    (    0 .    1    4   ;    0 .    1    9  –    0 .    9    9    )    0 .    4    0    (    0 .    1    8   ;    0 .    1    9  –    0 .    9    9    )    0 .    4    0    (    0 .    0    5   ;    0 .    3    7  –    0 .    4    9    )    N   o   w    i   t    h    b   e    l   o   w   t    h   e   r   a   p   e   u   t    i   c    l   e   v   e    l   a   t   a   n   y   t    i   m   e   a    f   t   e   r    i   n    i   t    i   a    l   a   c    h    i   e   v   e   m   e   n   t  —  —    3    1    2 Copyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 704–711.DOI: 10.1002/gps 708  a. macdonald  ET AL.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks