Most psychiatrists are familiar with the use of Amisulpride in the treatment of schizophrenia in adults. There is however currently no reliable evidence to guide antipsychotic prescribing for schizophrenia-like psychosis in the elderly (Essali and Ali, 2012). The ATLAS study (Howard et al, 2018) therefore proposed an innovative regimen for treatment of this very challenging, and often reluctant to engage population (Lam et al, 2016).

Very late onset of schizophrenia, as recognised by the International Late Onset Schizophrenia Group Consensus Statement (Howard et al, 2000) affects older adults over the age of 60. It is estimated that the prevalence of late-life schizophrenia at age 95 is 2.4%, (Royal College of Psychiatrists, 2012). Individuals who develop psychosis in this age range often experience suspicion, irritability, and ideas of reference, that progress into the development of visual and auditory hallucinations and delusional interpretation of life events (Zarit and Zarit, 2012).

Equally, with the progression in age it is not infrequent for this population to develop cognitive impairment, and in addition there is a lack of societal clarity between what is interpreted as “eccentric” or “psychotic”, which may confound the assessment and delay the treatment of psychosis (Royal College of Psychiatrists, 2012).

Very late-onset schizophrenia-like psychosis is not included as its own entity in the DSM-5 or ICD-10, however as life expectancy increases it seems that this health need may affect more people and the need for effective treatment may also become more urgent.

The Cochrane review of Antipsychotic drug treatment for elderly people with late onset schizophrenia (Essali and Ali, 2012) concluded that no evidence was available at the time to inform practice in this population. The ATLAS (Antipsychotic treatment of very late-onset schizophrenia-like psychosis) randomised, controlled, double blind trial, published in June 2018 in The Lancet Psychiatry, seems therefore a welcome and much needed initiative in this field.

Methods

ATLAS is a double blind, randomised controlled trial that comprised two phases:

• Stage 1: an initial 12-week period
• Stage 2: 24 weeks (later reduced to 12 weeks duration).

All participants were randomised into one of three groups:

1. Group “A” was allocated to treatment with Amisulpride 100 mg once daily in stages 1 and 2
2. Group “B” was assigned to Amisulpride 100 mg once daily in stage 1, followed by placebo in stage 2
3. Group “C” was randomised to placebo in stage 1 and Amisulpride 100 mg once daily in stage 2.

The original target was to recruit 300 participants in 25 centres in the UK, however this was later reduced to 100 due to challenges with recruitment.

The inclusion criteria included diagnosis of very late-onset schizophrenia-like psychosis (defined by International Consensus Group criteria, Howard et al 2000), with onset after the age of 60 years; BPRS score of ≥30 and capacity to provide consent.

Exclusion criteria included: MMSE score <25, primary diagnosis of affective disorder and serious physical health illness.

The primary outcomes were BPRS scores and medication discontinuation due to inefficacy. BRPS scores were assessed at weeks 4, 12 and 24 (or 36 for the participants enrolled to the original stage 2 duration of 24 weeks). The secondary outcomes were: extrapyramidal side effects (assessed by the Simpson and Angus Scale), compliance, quality of life (assessed by EuroQuol-5D, and WHO Quality of Life Scale) and resource utilisation (measured by Client Service, Receipt Inventory, reported separately).

Assessment results from participants who did not continue in the study after week 4 assessments were carried forward for final analysis.

Results

101 participants were consented into the study, and 92 were randomised into treatment. The mean age of participants was 80.2 years. BPRS scores averaged 41.3 at baseline. Thirty-three participants stopped treatment during stage 1, and further 25 during stage 2. Full data from all assessments was available for 34 participants.

• Amisulpride 100 mg was found to be significantly more effective in participants that completed the 12 weeks of treatment, versus participants in group C (placebo), with mean 11.9 points of improvement with amisulpride and 4.2 points in those on placebo
• In participants continuing amisulpride beyond the 12 week period, into stage 2, there was further reduction in BPRS score of 1.1 points, whilst the mean BPRS score of those who switched from amisulpride to placebo after 12 weeks increased by 5.2 points
• Participants in the placebo group were more likely to drop out due to perceived lack of efficacy than those on amisulpride (p<0.05)
• Neither of the groups noted statistically significant changes in quality of life
• Side effects, including serious adverse events were more frequent in the Amisulpride group:
  • Clinically significant increases in extrapyramidal side effects (EPSE) were noted in 11% of those taking Amisulpride (and none of those on placebo), three participants on amisulpride were hospitalised due to EPSE
  • Falls were also more frequent in the Amisulpride arm.

Conclusions

Despite the relatively small sample size and large drop out rates, the study provides evidence for the effectiveness of a very low dose of Amisulpride 100 mg in the treatment of very late onset schizophrenia-like psychosis. The use of Amisulpride had no effect on the quality of life measures, and there was significant increase in EPSE in participants receiving active treatment.

Strengths and limitations

Howard et al (2018) have delivered a study in a population that is known to be difficult to engage, and frequently abstains from taking prescribed antipsychotic medication (Sin Fai Lam et al, 2016).

Howard et al (2018) explained that data from participants in group C (switch from Placebo to Amisulpride) were not used for trial analysis in stage 2, due to lack of appropriate comparator group. This seems unfortunate, as the number of participants entering the trial was low and recruitment problematic, and therefore a different design may have resulted in a greater number and more power. For completeness it would appear important that the results from group C stage 2 are published. In addition, amendment of duration of stage two from 36 to 24 weeks suggests that dropout rates were originally underestimated. It seems that a pilot study may have been helpful in finalising the design of the study, and may have highlighted potential difficulties allowing the design to be optimised before the study opened for recruitment.

Participants who were taking another antipsychotic medication, but for whom change was appropriate were also enrolled into the trial. It is unclear however, how many participants that entered the trial were on active treatment for psychosis prior to coming into the study, and how many of these individuals were subsequently randomised to the placebo arm in group C. Stopping active treatment may have affected the BPRS scores in the Group C placebo arm, and therefore may have affected the interpretation of the gains against an active treatment with Amisulpride in groups A and B.

A further complicating factor is that participants were included in the study, with a MMSE score of 25 or above. Research suggests that a MMSE score of 25 in older adults, although it could be considered as “normal”, is also a strong predictor of people developing dementia in the next 3 years (Braekhus et al 1995). The data on MMSE score at the point of entry into the trial have been provided by the authors, however with no other cognitive tests being undertaken for assessment of cognitive function, and sensitivity of MMSE as a lone tool not providing 100% confidence in ruling our dementia (Creavin et al, 2016), one may wonder if it is possible that some participants in the ATLAS study may have experienced cognitive difficulties consistent with Mild Cognitive Impairment and be in prodromal stages of dementia. Secondly, a repeat MMSE score at the end of the study would reveal any changes and hence confirm whether the psychosis occurred independently of a decline in cognition.

Implications for practice

Most prescribers will be familiar with the recommended dose indicated for the treatment of schizophrenia in adults, the standard regimen being between 400mg and 800mg daily (BNF, 2018). Cardiovascular and metabolic abnormalities associated with prescribing of antipsychotic medication in the elderly population, especially those suffering from dementia, have led to recommendations of caution in prescribing this class of medication (BNF, 2018; DOH, 2009). Caution in prescribing may also have been related to the anticholinergic burden of many antipsychotic medications and to the negative effects on cognitions associated with their use (Montastruc et al, 2017).  When considering the anticholinergic burden however, Amisulpride appears to be a viable option, with low incidence of anticholinergic side effects (Juruena, et al 2010)

The efficacy of low dose Amisulpride in alleviating symptoms of psychosis in people with very late onset of schizophrenia-like psychosis, may encourage prescribers working with this population to offer this treatment option. The increase in extrapyramidal side effects and falls however suggests that the treatment is not free from side effects, and these have to be carefully monitored.

Conflicts of interest

None

Acknowledgements

The author(s) acknowledge support from the NIHR Oxford Cognitive Health Clinical Research Facility. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

Links

Primary paper

Howard R, Cort E, Bradley R, Harper E, Kelly L, Bentham P, Ritchie C, Reeves S, Fawzi W, Livingston G, Sommerlad A, Oomman S, Nazir E, Nilforooshan R, Barber R, Fox C, Macharouthu AV, Ramachandra P, Pattan V, Sykes J, Curran V, Katona C, Dening T, Knapp M, Gray R; ATLAS Trialists Group. (2018) Antipsychotic treatment of very late-onset schizophrenia-like psychosis (ATLAS): a randomised, controlled, double-blind trial. Lancet Psychiatry. 2018 Jul;5(7):553-563. doi: 10.1016/S2215-0366(18)30141-X. Epub 2018 Jun 4.

Other references

Department of Health (2009). The use of antipsychotic medication for people with dementia: time for action (PDF). A report for the Minister of State for Care Services by Professor Sube Banerjee.

Braekhus, A.; Laake, K.; and Engedal, K. (1995) A Low, ‘Normal’ Score on the Mini‐Mental State Examination Predicts Development of Dementia After Three Years, June 1995. Retrieved from https://doi.org/10.1111/j.1532-5415.1995.tb07201.

British National Formulary (online) London: BMJ Group and Pharmaceutical Press https://bnf.nice.org.uk. Accessed from the world wide web on the 27July2018.

Essali A, Ali G. Antipsychotic drug treatment for elderly people with late-onset schizophrenia. Cochrane Database of Systematic Reviews 2012, Issue 2. Art. No.: CD004162. DOI: 10.1002/14651858.CD004162.pub2.

Howard, R, Rabins, PV, Seeman, MV, Jeste, DV, and the International Late-Onset Schizophrenia Group (2000) Late-onset schizophrenia and very late-onset schizophrenia-like psychosis: an international consensus. Am J Psychiatry. 2000; 157: 172–178. [PubMed abstract]

Juruena, M. F., de Sena, E. P., & de Oliveira, I. R. (2010). Safety and tolerability of antipsychotics: focus on amisulpride. Drug, Healthcare and Patient Safety, 2, 205–211. http://doi.org/10.2147/DHPS.S6226

Lam, C.C., Reeves, S.J., Stewart, R., & Howard, R. (2016). Service and treatment engagement of people with very late-onset schizophrenia-like psychosis. British Journal of Psychiatry Bulletin, 40, 185-186. doi:10.1192/ pb.bp.115.051599

Montastruc, F.; Benevent, J.; Touafchia, A.; Chebane, L.; Araujo, M.; Guitton‐Bondon, E.; Durrieu, G.; Arbus, C.; Schmitt, L.; Begaud, B.; Montastruc, J. L. (2017) Atropinic (anticholinergic) burden in antipsychotic‐treated patients.  Fundamental & Clinical Pharmacology [27 Sep 2017, 32(1):114-119] doi: 10.1111/fcp.12321

National Institute of Clinical Excellence (2014) Psychosis and schizophrenia in adults. NICE guidelines. February 2014. http://www.nice.org.uk/guidance/CG178.

Rehse, M.; Bartolovic, M.; Baum, K.; Richter, D.; Weisbrod, M.; Roesch-Ely, D. (2016) “Influence of Antipsychotic and Anticholinergic Loads on Cognitive Functions in Patients with Schizophrenia,” Schizophrenia Research and Treatment, vol. 2016, Article ID 8213165, 10 pages, 2016. https://doi.org/10.1155/2016/8213165.

Royal College of Psychiatrists. (2012). Submission to the Schizophrenia Commission: Consultation response. Retrieved from https://www.rcpsych.ac.uk/pdf/RCPsych%20submission%20to%20the%20Schizophrenia%20Commission%20-%20June%202012.pdf

Sin Fai Lam, C. C., Reeves, S. J., Stewart, R., & Howard, R. (2016). Service and treatment engagement of people with very late-onset schizophrenia-like psychosis. BJPsych Bulletin, 40(4), 185–186. http://doi.org/10.1192/pb.bp.115.051599

Zarit, S.H. & Zarit, J.M. (2012). Mental disorders in older adults: Fundamentals of assessment and treatment. Guilford, UK: Guilford Press.

Photo credits

• Photo by Kiwihug on Unsplash
• Photo by Aris Sfakianakis on Unsplash
• Photo by Thiago Barletta on Unsplash