High doses of second-generation long-acting antipsychotics in the treatment of patients with severe resistant schizophrenia. A six-year mirror-image study.


Juan J. Fernández-Mirandaa,c, Silvia Díaz-Fernández a,b,c Francisco López-Muñoz,b,d,e,f ,

aAsturian Mental Health Service Área V- Servicio de Salud del Pº Asturias (SESPA), Gijón, Spain;

University Camilo José Cela, Faculty of Health Sciences, Madrid, Spain;

cAsturian Institute on Health Research (ISPA), Oviedo, Spain;

Hospital 12 de Octubre Research Institute, Neuropsychopharmacology Unit,Madrid, Spain;

Portucalense University, Portucalense Institute of Neuropsychology and Cognitive and Behavioural Neurosciences (INPP), Porto, Portugal;

Health Institute Carlos III, MICINN and FEDER, Thematic Network for Cooperative Health Research (RETICS), Addictive Disorders Network, Madrid, Spain.



Background: This study explores whether high-dose treatment with second-generation antipsychotic long-acting injectable (SGA LAI) may benefit patients with schizophrenia that are inadequately controlled on a standard dose. The objectives of this study have been to evaluate the retention, effectiveness and tolerability of high doses of formulations SGA LAI in the treatment of patients with severe resistant schizophrenia.

Methods: A 72-month observational, mirror-image study of patients with severe (Clinical Global Impression-Severity, CGI-S ≥ 5) resistant schizophrenia receiving treatment with ≥75 mg of risperidone long-acting injectable (RLAI) (N = 60), ≥175 mg of monthly paliperidone palmitate (PP) (N = 60), and ≥600 mg of aripiprazole once-monthly (AOM) (N = 30). All of the patients were previously treated with at least two different APs, with poor outcomes. Patients were eligible if deemed likely to benefit from treatment with SGA LAIs: at risk of medication non-compliance, with a lack of effectiveness, or adverse effects with previous APs. Assessment included the CGI-S, the WHO Disability Assessment Schedule (WHO-DAS), the Medication Adherence Rating Scale (MARS), laboratory tests, weight, adverse effects reported, reasons for treatment discontinuation, hospital admissions and suicide attempts.

Results: The average antipsychotic doses were: RLAI = 111.2 (9.1 SD) mg/14 days; PP = 231.2 (12.3 SD) mg eq./28 days; and AM = 780 (120 SD) mg/28 days.Tolerability was good for all LAIs, reducing the side effects reported and the changes in biological parameters compared to previous treatments, especially in the AOM group. Weight and prolactin levels decreased in all LAI treatments; the reduction was statistically significant only among patients treated with AOM (p < 0.05). Two patients discontinued treatment due to side effects with AOM, five with PP and nine with RLAI. There were four discontinuations with RLAI, two with PP, and one with AOM due to a lack of effectiveness (severe symptoms or hospital admission). After three years, the scores decreased in CGI-S (p < 0.01) and in WHO-DAS in the four areas with all injectables. MARS increased with all LAIs (p < 0.01), especially with PP and AOM. We have reported a statistically significant decrease in both hospital admissions (p <0.001) and suicide attempts (p < 0.001) at the end of 36-month treatments, compared to the previous three years, without any difference across the three LAIs. In the previous three years, 60 patients discontinued their AP treatment, and 11 during the three-year follow-up (p < 0.0001).

Conclusions: Our study indicates the good effectiveness and tolerability of RLAI, PP and AM at high doses. These SGA LAI treatments improved treatment adherence and outcomes (hospitalizations, suicide attempts, clinical severity, disability) of the patients, with good tolerability, helping them to achieve clinical stabilization and better functioning. Therefore, we suggest that, in some illness critical conditions, high doses of SGA LAIs could represent an alternative to clozapine, to date the only therapy for resistant schizophrenia

Keywords: schizophrenia, long-acting antipsychotic, dose, adherence, effectiveness, tolerability.




High-dose therapy is defined as doses that exceed the approved recommended dose (as stated by the European Medicines Agency). Treatment-resistant schizophrenia is usually defined as an illness that has recorded an insufficient response (no significant clinical improvement) to appropriate trials in terms of the dose (equivalents to 1000 mg/day of chlorpromazine), duration (at least 3 periods of 6 weeks in the previous 5 years) and adherence of at least two different antipsychotic (AP) medications (1). Guidelines indicate that at least one of the drugs should be a non-clozapine second-generation antipsychotic (SGA) (2). Some studies estimate that 30% of people with schizophrenia have a poor response to antipsychotic (AP) medications. Clinical trials show that only 20% of patients with schizophrenia present complete remission with appropriate antipsychotic treatment and 20%-30% of this group suffer a relapse during the first year of treatment (3).

Clozapine is commonly recommended for treating patients that are inadequately controlled following trials with two different classes of Aps 4 , and it is the only AP approved for treatment-resistant schizophrenia (TRS); but 30% of TRS patients experience little or no benefit from it (4,5). SGAs, which are safer in some respect than clozapine, have also been reported to be effective in subgroups of TRS patients. Response to these APs require attention to the issue of duration of treatment and dose (6,7).

Due to safety concerns over clozapine, as well as a lack of response in a significant number of patients, treatment with a high-dose SGA is often applied, (8,9) even though these strategies are not recommended in most current guidelines on clinical practice (10–12). High-dose usage ranges widely from approximately 5% to 50%, and is likely to be at the higher end for patients with schizophrenia that are inadequately controlled on standard-dose AP monotherapy (12,13).

Not only the effectiveness but also the tolerability of APs is important for reinforcing treatment compliance, and consequently achieving rehabilitation goals in people with severe schizophrenia. The rate of nonadherence to treatment among people diagnosed with schizophrenia has been estimated at between 20%-56%, and may contribute to relapses, hospital admissions, loss of function, and suicidality.1,8 Several factors may affect treatment adherence in patients with schizophrenia, including AP medications (lack of efficacy, side effects, frequency of administration, duration of treatment, etc.) (14,15).

 It has been suggested that SGAs are more effective than first-generation ones (FGAs) in the treatment of negative, cognitive, and depressive symptoms in patients with schizophrenia, and that they are also better tolerated than FGAs (11,16,17). On the other hand, long-acting injectable (LAI) APs are prescribed to reduce nonadherence and relapse in people diagnosed with schizophrenia, and can be considered an effective treatment strategy for improving adherence. Clozapine and second-generation LAIs have recorded the highest rates of treatment adherence and relapse prevention in schizophrenia patients (14,16,18,19). Several meta-analyses have demonstrated their superiority over oral APs (OAPs) for preventing hospitalisation (20,21). Cohort studies find mixed results, but most of them report better results for LAIs than for OAPs, (22) as do mirror-image studies (23–27).

The current debate focuses especially on the use of OAPs or LAI APs for improving adherence and treatment outcomes in schizophrenia. In general, the use of SGA LAIs has already been recommended (18,28) although there is still no clear consensus: meta-analyses of randomised clinical trials (RCTs) comparing LAIs with OAPs (both SGAs) have provided contrasting results (20,21). These results are not consistent because RCTs are heavily influenced by the biases of the study design (16,20,21). The superior effectiveness of LAIs over OAPs is more evident in mirror-image studies and in cohort studies (17,19,22,25). Certain recent naturalistic studies even report greater tolerability, fewer relapses, and less suicidal behaviour with high doses of LAI SGAs in patients with severe schizophrenia (8,27).

Clinicians treating schizophrenia often prescribe high doses, with a common reason being a poor response to standard treatment (8,13,14,18). For most APs, both first- and second-generation ones, a near maximal effective dose is lower than the maximum licensed dose (29,30). Most clinical guidelines and meta-analyses do not provide any convincing evidence that an AP dosage higher than the maximum licensed dose is more effective than the standard dosage for treatment-resistant schizophrenia.

The CADTH Optimal Use Report considers the value of high-dose SGA versus standard-dose non-clozapine APs for treatment-resistant schizophrenia, and interpret the data available as a lack of evidence of any advantages (10). The Royal College of Psychiatrists states that ‘there does not seem to be any justification for the use of high-dose antipsychotic medication for relapse prevention in schizophrenia’(12).

However, there is a reasonable amount of evidence supporting the rationale behind prescribing high doses: an insufficient amount of drug might reach the effect site because of patients’ individual differences in pharmacokinetics, as well differences at the effect site in some patients (pharmacodynamic differences) (13,29).

This study explores whether high-dose treatment with SGA LAIs may benefit patients with schizophrenia that are inadequately controlled on a standard dose. The objectives of this study have been to evaluate the retention, effectiveness and tolerability of high doses of SGA LAI formulations in the treatment of patients with severe resistant schizophrenia.




With a 36-month perspective, an observational, mirror-image (36-month retrospective) study has been conducted of patients (N = 150; 6-year follow-up) with severe resistant schizophrenia (Clinical Global Impression-Severity, CGI-S, of five and over), treated in a community-based programme, which for clinical stabilisation underwent treatment with 75 mg and over of RLAI (N = 60), 175 mg and over of monthly PP (N = 60), and 600 mg and over of AOM (N = 30). All of the patients were previously treated with at least two different APs (at least one of them SGA), with poor outcomes.


Patients were aged 18 and over, and were eligible if deemed likely to benefit from treatment with SGA LAIs: at risk of medication non-compliance, with a lack of effectiveness, or adverse effects with previous APs. Exclusion criteria were  mental retardation or moderate/severe cognitive impairment. The subjects enrolled in the study were the first ones (among all those treated with SGA LAIs) that needed treatment with doses over the standard ones for clinical stabilisation due to a lack of effectiveness (CGI-S ≥ 5 or hospital admission due to clinical decompensation after three months of therapy with standard doses). So, as study population, individuals were chosen using criteria related to the characteristics of the research, for ease of convenience (non-probabilistic sampling), and sampling error cannot be calculated. The recruitment period ran from January 2014 to December 2016.

The mean age at the start of the high-dose treatment was 43.5 years (standard deviation, SD: 9.8). 60.7% were men and 39.3% women. Table 1 reports the patients’ clinical variables (years of illness and previous treatment; the number of prior instances of treatment discontinuation, psychiatric hospitalisations and suicide attempts; the kind of therapy [oral or other LAIs, FGA or SGA] before the switch to the three LAIs; and the reasons for the switch [non-adherence to treatment, lack of efficacy, or intolerable adverse effects]).


Measurement tools

The assessment of effectiveness included the CGI-S (Severity of illness), the WHO Disability Assessment Schedule – WHO-DAS (four areas on functioning and disability: self-care, occupational, family, and social impairment) and the Medication Adherence Rating Scale, MARS (adherence, satisfaction and insight on the need for medication) at the beginning and after 12, 24 and 36 months of therapy. All the causes of treatment discontinuation, psychiatric hospital admissions and documented suicide attempts in the previous three years and during the follow-up were recorded. Other clinicians besides those treating the patients were part of the study assessing changes in CGI-S and WHO-DAS.

Drug tolerance was monitored annually through laboratory tests (haematology, biochemistry, and prolactin levels), with weight and any adverse effects being reported at each three-monthly visit (no specific scales were used).

All psychopharmacological and anti-Parkinsonian medications were registered in the previous three years and during the follow-up.

The primary outcome was failure in effectiveness, which reflected inadequate control of the schizophrenia symptoms and the lack of current AP therapeutic benefit with CGI-S ≥ 5 or hospital admissions.

The secondary outcome was tolerability, which included changes in weight from the baseline and changes for the worse in glucose, cholesterol, triglyceride, and prolactin. Other important secondary outcomes included measures of abnormal involuntary movements, akathisia, and Parkinsonism. The weight and measures of neurological or other adverse effects were recorded on all the study visits (every three months). Laboratory blood tests were conducted at a basal visit and at months 12, 24 and 36.

Statistical analysis

The main statistical analyses involved comparing treatment discontinuation, hospital admissions, suicide attempts, scale scores and laboratory test results before and after high-dose treatments. Chi2 was used for qualitative variables, and McNemar’s test was used to compare paired proportions. A Student’s t-test for paired data was used for quantitative variables. The confidence interval was established at 95%. The “R Development Core Team” program (3.4.1 version)-Package MASS (7.3-45 version) was used for data processing.


The study was carried out in accordance with the Code of Ethics of the WMA ethical principles (Declaration of Helsinki), and was approved by the Ethical Clinical Research Committee of the Asturian Health Service (Comité de Ética de Investigación con medicamentos (CEIm), R 1090/2015. Hospital Universitario Central de Asturias N-1, S3.19). All the patients (or their legal representatives, if appropriate) signed informed consent forms to begin their treatment on the programme; and were informed about AP doses higher than the recommended ones, to which they consented.



We have reported the following average doses of the three LAIs: RLAI = 111.21 (±9.12 SD) mg/14 days; PP = 231.23 (±12.32 SD) mg eq./28 days; AOM = 780.08 (±120.12 SD) mg/28 days.

Tolerability was good for all LAIs, reducing the side effects reported and the changes in biological parameters compared to previous treatments, especially in the AOM group. Weight and prolactin levels decreased in all LAI treatments; the reduction in the number of subjects with altered prolactin levels was clearly marked among patients treated with AOM (from five to only one). Two patients discontinued treatment due to side effects with AOM (Akathisia), five with PP [Extrapyramidal symptoms (EPS)/Parkinsonism (3), hyperprolactinaemia (1) and sedation (1)], and nine with RLAI [EPS/Parkinsonism (4), hyperprolactinaemia (1), sedation (3) and hyperlipemia (1)]. Adverse effects reported, weight and laboratory tests are shown in Table 2.                                               

There were four discontinuations with RLAI, two with PP, and one with AOM due to a lack of effectiveness (severe symptoms or hospital admission).

CGI-S, WHO-DAS, MARS scores, hospital admissions and suicide attempts in the 36-month follow-up are reported in Table 3. Subjects included in this report showed a global improvement in all the scales. After three years, the scores decreased in CGI-S (t (60) = 2.7, p < 0.01) and in WHO-DAS in the four areas (self-care (t (60) = 12.54, p <  0.005; occupational (t (60) = 8.22, p < 0.01; family (t (30) =10.06, p <  0.01; and social impairment (t (60)=9.77,  p < 0.05) with all injectables. MARS increased with all LAIs (t (150) = 17.02,  p < 0.01), especially with PP and AOM .

We have reported a statistically significant decrease in both hospital admissions (1.1 ±1.2 SD vs. 0.3±0.2 SD (t (75) = 14.87,  p < 0.001) and suicide attempts (0.8±0.6 SD vs. 0.2±0.2 SD (t (75) = 5.36, p < 0.001) at the end of 36-month treatments, compared to the previous three years, without any difference across the three LAIs.

Significant statistical differences can be seen at the first year of the follow-up.

In the previous three years, 60 patients (40%) discontinued their AP treatment, and 11 during the three-year follow-up (x 2 (1, N = 150) = 16.47, p < 0.0001) .

There was a reduction in anti-Parkinsonian, antidepressant and anxiolytic treatments in comparison to previous AP therapy, and between the three LAIs (p < 0.01 for PP and AOM).

Forty-six patients (30.6%) were previously treated with two APs, one on LAI at high doses, and thirty-one (21%) on doses that exceeded the licensed ones (Table 1).

After 36 months, 47 out of 60 patients (78.3%) continued with RLAI; 53 out of 60 (88.3%) with PP, and 27 out of 30 (90%) with AOM.

Gender and age were not related to AP treatment outcomes (adherence, effectiveness and tolerability). No significant differences were found between men and women in percentage of treatment discontinuation (8.1% vs. 8.3%), and mean of hospital admissions (0.3 vs 0.2) and suicide attempts (0.2 vs. 0.2). We have reported side effects or alterations in biological parameters in 30.4 % of subjects (31.9% in men and 29.2% in women).



Rationale behind prescribing high doses

From a pharmacological perspective, there are two main reasons higher doses of APs might be theoretically justified in some cases: insufficient drug might reach the effect site (pharmacokinetic differences in individual patients), and the nature of the effect site itself (pharmacodynamic differences). There is sufficient evidence to support the first rationale for high-dose prescriptions: pharmacokinetic differences mean low drug plasma levels and an insufficient AP blockade of D2 receptors at standard doses in some patients; with regard to the second rationale, there is currently little evidence to support differences in D2-receptor levels or function in some patients requiring the use of high doses (12,29).

Many attempts have been made to confirm a dose–response relationship for APs using plasma levels. The apparent positive dose-response relationship we see with these higher doses could be due to pharmacokinetic issues, and the lack of blood levels information is a deficiency. People with low blood levels are out there and they have bad treatment outcomes (31). Plasma levels quantification (‘therapeutic drug monitoring’) is considered the best way for precision medicine with antipsychotic usage and that it is needed to guide practice (32) and to augment antipsychotic use and dosing decisions. And it can be helpful in especial when there is a lack of therapeutic response, relapse, or adverse drug reactions, offering a method to improve safety and efficacy (33). However, some authors stress that what is important is drug concentration in the brain rather than dose or drug plasma level (12,30).

On the other hand, although in treatment-resistant schizophrenia clozapine is considered the standard treatment, in a meta-analysis with forty blinded RCTs with 5172 participants, the most surprising finding was that clozapine was not significantly better than most other drugs34. Insufficient evidence exists on which antipsychotic is more efficacious for patients with treatment-resistant schizophrenia, and RCTs provide little evidence of the superiority of clozapine compared with other SGAs. In this sense, future clozapine and other SGAs  studies with high doses and patients with extremely treatment-refractory schizophrenia are needed to change the current evidence 3,7).

The main reasons for the switch in our study were a lack of effectiveness and non-adherence to treatment. The results show a remarkable improvement in both treatment adherence and effectiveness after changing to a LAI-SGA formulation at high doses, as well as the redundancy of

AP polytherapy.

Treatment effectiveness and LAI SGA high-dose therapy

Although studies of relapse prevention do not usually test high-dosage regimes, the Canadian guidelines recommend that high doses of non-clozapine SGA ‘should not be used instead of standard doses in patients with schizophrenia that manifest an inadequate response to a standard-dose antipsychotic (10). Similarly, the schizophrenia treatment guidelines issued by the World Federation of Societies of Biological Psychiatry conclude that there is ‘no good evidence that high maintenance doses are more effective in preventing relapse than standard doses (11). Finally, the Royal College of Psychiatrists states that ‘there is no convincing evidence base to support the prophylactic value of continuing the higher dose of antipsychotic medication long term or the prescription of further dose increments (12). The CADTH review does not find any significant differences in efficacy between high-dose non-clozapine SGA treatment and standard-dose clozapine, with the exception of Global Assessment of Functioning (GAF) scores, which improve less in the high-dose SGA group (10). When high-dose non-clozapine SGAs have been compared to standard-dose non-clozapine APs, meta-analyses do not detect any significant differences.

In general, LAIs are considered appropriate for maintenance treatment in patients with stabilised schizophrenia, (16,18,21,22,28) and they seem to reduce relapses, re-admissions, (16,18–22) and indirectly, suicidal behaviour (15,26).

Although guidelines, on the basis of evidence reviews, conclude that higher doses of antipsychotics are not more effective than standard doses, and in general should be avoided, there is some evidence that the average doses of SGA needed for improvement in psychopathology in TRS patients are significantly higher than those for non-TRS patients35,36. On the other hand, other studies of TRS patients, one with long-acting injectable risperidone (50 vs 100 mg 4 times a day for 2 weeks) did not find efficacy to be dose dependent. It is possible that the variability of dose-response relationships among SGAs is related to differences in their pharmacokinetics, metabolism, and relative differences in affinities for a variety of receptors which have been shown to have a role in psychopathology and response to APs (37). The absence of D4  dopamine receptor antagonism may be important to the efficacy of SGAs other than clozapine. Also, higher doses of some SGAs may produce greater D2  dopamine receptor blockade, which may limit the beneficial effects of 5-HT2A receptor blockade (7).

Studies with SGAs in TRS patients also found time-dependent improvement in psychopathology (37). Some studies indicate that SGAs with pharmacology similar to that of clozapine benefit many but not all TRS patients, and that time to response is often delayed by several months compared with non TRS patients (34). Meta-analysis of randomized controlled trials of clozapine versus other comparators concluded that clozapine was superior for total symptoms, but only in the short term, not the long term (7,38).

Hospital admissions (considered as relapses) are markedly lower in this research than earlier ones enrolling people at increased risk of nonadherence and relapse. Our study on assessing the effectiveness of high-dose atypical antipsychotic therapy in patients with schizophrenia inadequately controlled on standard-dose antipsychotics has found significant improvements favouring high-dose SGA (RLAI, PP, and AOM) treatment strategies, recording not only a high treatment adherence but also a decrease in illness severity (lower GCI-S scores), hospital admissions and suicide attempts. Patients with these high AP doses were clinically stabilised and achieved better social functioning levels (lower WHO-DAS scores) and increased their awareness of the illness (and AP medication adherence considering MARS scores). Moreover, they needed less anti-Parkinsonian, antidepressant and anxiolytic concomitant treatment.

Tolerability and LAI SGA high-dose therapy

Although the likelihood and intensity of most of the adverse effects of APs increase with dosage, the risk also increases in step with the speed of dosage. Some reactions are unpredictable in these cases, and might reflect individual patient susceptibilities; others are neither clearly idiosyncratic nor dose-related. There is a need to consider reactions that are either clearly or potentially dose-related, although it should be acknowledged that several factors may confound dose–response relationships (drug pharmacokinetics, role of metabolites, and patient’s clinical characteristics) (12,29,39,40).

There is still a lack of robust data on the relative risk of SGA’s adverse effects within the standard dosage range, and much more so at high dosages. Although some studies find that more patients experience EPS and elevated prolactin in high-dose non-clozapine SGA treatment than with standard-dose clozapine treatment, others report that fewer patients had Parkinsonism on a high-dose SGA treatment compared to standard-dose clozapine, and the rate of withdrawals due to adverse events is also lower in high-dose SGA patients (10,29).

In relation to gender factors and high-doses, pharmacokinetic considerations alone would suggest that the threshold for determining what is a high dose of APs for women should be scaled down. In terms of clozapine, women require a dose that is around 20% less than men to achieve the same plasma clozapine level. A general conclusion is that women are likely to require lower doses than men, including high-dose scenarios (40).

On balance, the data available do not provide conclusive evidence to support the safety of high-dose treatment strategies. Observational evidence is similarly inconclusive (27,41–43). As the evidence regarding harm is inconclusive, the risk of all these potentially adverse effects needs to be weighed against the benefits of receiving effective treatment (18,19,21,42,43).

In our study, the tolerability of high doses of LAI SGA (for PP and, especially, AOM) is very good, with very few and mild side effects reported, especially compared to previous therapy. This is useful for improving treatment adherence, and thus helping to achieve clinical stabilisation. The safety of high doses is high according to laboratory test results, and even weight and prolactin levels decrease compared to previous AP treatments.

Contrary to expectations, no increase in EPS has been found. In fact, there is a decrease in the use of anti-Parkinsonian medication. The patients in this study have recorded few EPS after receiving these high doses. Some of them may have been rapid metabolisers and had low levels of APs on whatever they were taking previously, and with the high dose SGA LAI they had therapeutic doses and did well clinically, while not having many EPS. There have been few other major side effects. The number of patients with sedation has decreased, probably helping to improve functioning.

Low discontinuation rates also support patients’ good acceptance of an injectable AP formulation. The increase in MARS scores could also be a result of this good tolerability. However, previous studies indicate that treatment satisfaction is positively associated with symptom improvement, which may also be related to compliance. Neither movement disorders nor prolactin elevation have proven to predict medication satisfaction (8,14,15,44).

The high doses in these patients might also have been better tolerated due to pharmacokinetic or pharmacodynamic issues. Severely ill patients might tolerate high levels of SGA without EPS or other side effects; patients seem in many cases to not get EPS from high doses of risperidone or paliperidone. Assuming blood brain penetration (in our study there were no PET scan data to confirm D2 occupancy), it has long been observed that such patients exist (“Type 3 neuroleptic resistance”) (45), and our population may have a large number of them. But, we still do not know what the best treatment is for them. Yet, these patients are in fact exceptions to the usually well-demonstrated dose-response relationship with EPS.

The prolactin levels winding up a little lower on RLAI and PP after three years is not a surprise because studies have shown that some tolerance gradually develops over months to the elevated prolactins (and here we have a 3-year period that is involved). There was an expected large decrease in prolactin on aripiprazole LAI; plenty of evidence show this due to its “prolactin sparing” partial D2 agonist effect (46).

Study strengths and limitations

Our study’s main strength is that it is the first to assess the treatment retention, tolerability and effectiveness of high doses of SGA LAIs in a broad real-world sample (N = 150) and for a long period (36 months). This allows us to measure retention in people with severe treatment-resistant schizophrenia, one of their main issues, and link it to satisfaction with high-dose pharmacological treatment due to their high tolerability and effectiveness. We highlight that our study has been based on routine clinical practice involving patients with a lack of awareness of the illness that have abandoned their previous therapy in a significant percentage, thereby helping to confer upon the study a high external validity.

Although more serious symptoms do not automatically mean higher doses, they have been more severe in this sample than in the clinical trials.

Several limitations need to be addressed.

We have designed an open-label non-randomised study under pragmatic conditions. There is no control group, and there is no active comparator. The study’s design with no control group is a limitation that may mean our results have lower internal validity. We have designed a mirror-image study in order to overcome this limitation.

The study has not prospectively defined the onset of a clinically meaningful change. We have used the CGI-S as a generally acceptable measure of a significant change in severity. It is an unspecific instrument, and may thus constitute a limitation. Other medical staff apart from the treating clinicians have been involved in the study, assessing changes in CGI-S and WHO-DAS in order to avoid the risk of a desirability effect.

The study does not include a comparison with OAP medications. Nevertheless, the use of LAI AP medications is supported by a number of systematic reviews and expert panels for outpatients with a higher risk of relapse. No formal side-effect assessment scales have been applied. No blood levels have been recorded, so we do not know the relationship between this level and clinical response. We assume that the study would have been more valuable with this additional information: although the plasma levels of LAIs are highly variable, the reasons for this are unclear. It would add some clarity to this question and reveal levels at which these severely ill patients improve, and what side effects they might have at those levels. 

RCTs with SGA LAIs are not specifically designed to examine a particularly ill patient population. However, all the patients in our study have been rated as severely ill by the CGI-S at the study centre. The results presented here cannot be generalised to patients that are not severely ill due to the possible biases inherent to the study’s design, and further studies are required to confirm our findings.

We point out that many patients in this study were before on high doses or in combination of other APs, as shown in table 1, and they were not doing well on these medications (mostly oral). They were mostly non-adherent or non-responsive, and only a few were intolerant. The consequences of using these doses could be causing up regulation of the receptor systems they are blocking with the result that some patients may continue to need higher doses in order to not destabilize even further. Other ones needed high doses after three months on standard ones and with lack of effectiveness. As there was no control group getting normal doses of LAIs, we considered these previous standard doses treatment as a sort of comparison. We also stress that all the patients received the same psychosocial treatment, so they did not benefit just from being in the study, getting the extra attention involved.

A final point is that the mean doses of 175 for PP and 780 for AO are actually not outside the upper end of usual (234 for PP, 800 for AOM) in some countries, but they are in Spain.

We stress that it not possible to reach definitive conclusions because of the study’s observational design, a sample size that is not big enough to report different LAIs, and the lack of LAI blood level testing (or neuroimaging tests such as PET or SPECT to highlight D2 receptor occupancy). Nevertheless, this study has contributed by highlighting the need to consider new treatment strategies for severe and treatment-resistant schizophrenia.



This study assesses the safety, tolerability, and effectiveness of high-dose atypical AP therapy, assured by the use of LAIs, avoiding possible and uncontrolled poor compliance with the use of OAPs, in patients with schizophrenia inadequately controlled on standard-dose APs. Significant improvements have been found favouring high-dose SGA treatment strategies.

In terms of effectiveness, the patients have recorded not only a high treatment adherence but also a decrease in illness severity, disability, hospital admissions and suicide attempts compared to previous treatments at standard doses and with oral formulations. The patients have also recorded clinical stabilisation, better functioning levels, and increased awareness of the illness with high AP doses.

In terms of safety, no clinically significant differences are evident between a high-dose therapy compared to a standard-dose one. In fact, fewer side effects and laboratory test alterations have been reported than in previous treatments. The tolerability of high doses of second generation long-acting antipsychotics (RLAI, and especially for PP and AOM) is very good for improving treatment adherence in patients with severe treatment-resistant schizophrenia. Low discontinuation rates also support patients’ good acceptance of an injectable antipsychotic formulation.

These patients were clozapine candidates in routine clinical practice. It is not possible, however, to state that a high dose of SGA LAIs could be an alternative to clozapine treatment, as no comparison has been made. It is possible that there is a subgroup of patients who need high dose LAIs and can avoid clozapine that way, but more study is needed. Surely, what treatment-resistant patients should get, before clozapine, is a trial, with blood levels, of regular dose LAI which ensures adherence and absorption into the plasma, at least, if not into the brain. And get these tolerable high doses if necessary after that. These LAIs are extremely costly, so they should not be used at these doses if it is not necessary.

Anyway, clinicians could consider them to a be more adherent and tolerable treatment option in the management of severely symptomatic patients. We suggest the need for more research studies to find new strategies to treat patients with severe treatment-resistant schizophrenia, for whom to date only clozapine is indicated. In general terms, the widespread implementation of high doses of long-acting SGA could be an option for people with schizophrenia with severe symptoms and impairment, and at high risk of abandoning the treatment.

Longer-term studies of sufficient methodological quality and sample size are required to determine more accurately whether high-dose treatment strategies have clinical value in the treatment of patients with severe treatment-resistant schizophrenia. Further quality studies are called for to more precisely assess the effectiveness and safety of high doses of non-clozapine SGA LAIs.



1.        Molina JD, Jiménez-González AB, López-Muñoz F, Cañas F. Evolution of the concept of treatment-resistant schizophrenia: Toward a reformulation for lack of an adequate response. J Exp Clin Med. 2012;4(2):98-102. doi:10.1016/j.jecm.2012.02.003

2.        National Institute for Health and Clinical Excellence. Schizophrenia: Core interventions in the treatment and management of schizophrenia in primary and secondary care (update). 2009;(82):London: Nice. doi:10.1093/schbul/sbp021

3.        Molina J de D, Quintero J, García-Laredo E, López-Muñoz F, Correas-Lauffer J, Barbudo E, et al. Cognitive effects of combined amisulpride and quetiapine treatment in patients with refractory schizophrenia : A Naturalistic, Prospective Study. Am J Ther. 2019; 27(4):e346-e355. doi:10.1097/MJT.0000000000000956

4.        López-Muñoz F, Sanz-Fuentenebro J, Rubio G, García-García P, Álamo C. Quo Vadis clozapine? A bibliometric study of 45 years of research in international context. Int J Mol Sci. 2015;16(9):23012-23034. doi:10.3390/ijms160923012

5.        Meltzer HY. Clozapine. Clin Schizophr Relat Psychoses. 2012;6(3):134-144. doi:10.3371/CSRP.6.3.5

6.        Hasan A, Falkai P, Wobrock T, Lieberman J, Glenthoj B, Gattaz WF, et al. World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for Biological Treatment of Schizophrenia, Part 1: Update 2012 on the acute treatment of schizophrenia and the management of treatment resistance. World J Biol Psychiatry. 2012;13(5):318-378. doi:10.3109/15622975.2012.696143

7.        Meltzer HY, Share DB, Jayathilake K, Salomon RM, Lee MA. Lurasidone improves psychopathology and cognition in treatment-resistant schizophrenia. J Clin Psychopharmacol. 2020;40(3):240-249. doi:10.1097/JCP.0000000000001205

8.        Fernández-Miranda JJ, Caramés-García V, Sánchez-García A. Effectiveness, good tolerability, and high compliance of doses of risperidone long-acting injectable higher than 75 mg in people with severe schizophrenia: A 3-year follow-up. J Clin Psychopharmacol. 2015;35(6):630-634. doi:10.1097/JCP.0000000000000400

9.        Sommer IE, Begemann MJH, Temmerman A, Leucht S. Pharmacological augmentation strategies for schizophrenia patients with insufficient response to clozapine: A quantitative literature review. Schizophr Bull. 2012;38(5):1003-1011. doi:10.1093/schbul/sbr004

10.      CADTH. Systematic review of combination and high dose AAPs for schizophrenia. 2011;1(1). https://www.cadth.ca/sites/default/files/pdf/H0503_AAP_science-report_e.pdf.

11.      Hasan A, Falkai P, Wobrock T,  Lieberman J, Glenthoj B, Gattaz WF, et al. World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for Biological Treatment of Schizophrenia, Part 2: Update 2012 on the long-term treatment of schizophrenia and management of antipsychotic-induced side effects. World J Biol Psychiatry. 2013;14(1):2-44. doi:10.3109/15622975.2012.739708

1 2.     Royal College of Psychiatrists. Consensus statement on high-dose antipsychotic medication. R Coll Psychiatr. 2014;(October 2005):1-53.

13.      Mace S, Taylor D. Reducing the rates of prescribing high-dose antipsychotics and polypharmacy on psychiatric inpatient and intensive care units: results of a 6-year quality improvement programme. Ther Adv Psychopharmacol. 2015;5(1):4-12. doi:10.1177/2045125314558054

14.      Kane JM, Kishimoto T, Correll CU. Non-adherence to medication in patients with psychotic disorders: epidemiology, contributing factors and management strategies. World Psychiatry. 2013;12(3):216-226. doi:10.1002/wps.20060

15.      Leucht S, Cipriani A, Spineli L, Mavridis D, Orey D, Richter F, et al. Comparative efficacy and tolerability of 15 antipsychotic drugs in schizophrenia: A multiple-treatments meta-analysis. Lancet. 2013;382(9896):951-962. doi:10.1016/S0140-6736(13)60733-3

16.      Lai Ming Hui C, Wing Yan Poon V, Tung Ko W, Ho Yee M, Wing Chung C, Ho Ming Lee E, et al. Risk factors for antipsychotic medication non-adherence behaviors and attitudes in adult-onset psychosis. Schizophr Res. 2016;174(1-3):144-149. doi:10.1016/j.schres.2016.03.026

17.      Leucht C, Heres S, Kane JM, Kissling W, Davis JM, Leucht S. Oral versus depot antipsychotic drugs for schizophrenia-A critical systematic review and meta-analysis of randomised long-term trials. Schizophr Res. 2011;127(1-3):83-92. doi:10.1016/j.schres.2010.11.020

18.      Nasrallah HA. Triple advantages of injectable long acting second generation antipsychotics: Relapse prevention, neuroprotection, and lower mortality. Schizophr Res. 2018;197(2018):69-70. doi:10.1016/j.schres.2018.02.004

19.      Taipale H, Mehtälä J, Tanskanen A, Tiihonen J. Comparative effectiveness of antipsychotic drugs for rehospitalization in schizophrenia – A nationwide study with 20-year follow-up. Schizophr Bull. 2018;44(6):1381-1387. doi:10.1093/schbul/sbx176

20.      Kishimoto T, Robenzadeh A, Leucht C, Leucht S, Watanabe K,  Mimura M, et al. Long-acting injectable vs oral antipsychotics for relapse prevention in schizophrenia: A meta-analysis of randomized trials. Schizophr Bull. 2014;40(1):192-213. doi:10.1093/schbul/sbs150

21.      Misawa F, Kishimoto T, Hagi K, Kane J M, Correll CU. Safety and tolerability of long-acting injectable versus oral antipsychotics: A meta-analysis of randomized controlled studies comparing the same antipsychotics. Schizophr Res. 2016;176(2-3):220-230. doi:10.1016/j.schres.2016.07.018

22.      Tiihonen J, Mittendorfer-Rutz E, Majak M, Mehtälä J, Hoti F, Jedenius E, et al. Real-world effectiveness of antipsychotic treatments in a nationwide cohort of 29 823 patients with schizophrenia. JAMA Psychiatry. 2017;74(7):686. doi:10.1001/jamapsychiatry.2017.1322

23.      Di Lorenzo R, Cameli M, Piemonte C, Bolondi M, Landi G, Pollutri G, et al. Clinical improvement, relapse and treatment adherence with paliperidone palmitate 1-month formulation: 1-year treatment in a naturalistic outpatient setting. Nord J Psychiatry. 2018;72(3):214-220. doi:10.1080/08039488.2017.1420819

24.      Taylor DM, Sparshatt A, O’Hagan M, Dzahini O. Effect of paliperidone palmitate on hospitalisation in a naturalistic cohort – a four-year mirror image study. Eur Psychiatry. 2016;37:43-48. doi:10.1016/j.eurpsy.2016.04.009

25.      Díaz-Fernández S, Frías-Ortiz DF, Fernández-Miranda JJ. Mirror image study (10 years of follow-up and 10 of standard pre-treatment) of psychiatric hospitalizations of patients with severe schizophrenia treated in a community-based, case-managed programme. Rev Psiquiatr Salud Ment. 2019;(xx). doi:10.1016/j.rpsm.2019.04.004

26.      Díaz-Fernández S, Frías-Ortiz DF, Fernández-Miranda JJ. Suicide attempts in people with schizophrenia before and after participating in an intensive case managed community program: A 20-year follow-up. Psychiatry Res. 2020;287(December 2018):112479. doi:10.1016/j.psychres.2019.112479

27.      Fernández-Miranda JJ, Díaz-Fernández S. Tolerability of effective high doses of paliperidone palmitate in patients with severe resistant schizophrenia. Int Clin Psychopharmacol. 2017;32(1):6-12. doi:10.1097/YIC.0000000000000151

28.      Citrome L, Reist C, Palmer L, Montejano L, Lenhart G, Cuffel B, et al. Dose trends for second-generation antipsychotic treatment of schizophrenia and bipolar disorder. Schizophr Res. 2009;108(1-3):238-244. doi:10.1016/j.schres.2008.11.017

29.      Correll CU, Citrome L, Haddad PM, Lauriello J, Olfson M, Calloway SM, et al. The use of long-acting injectable antipsychotics in schizophrenia. J Clin Psychiatry 2016:1-24. doi:10.4088/JCP.15032su1

30.      Gardner DM, Murphy AL, O’Donnell H, Centorrino F, Baldessarini RJ. International consensus study of antipsychotic dosing. Am J Psychiatry. 2010;167(6):686-693. doi:10.1176/appi.ajp.2009.09060802

31.      McCutcheon R, Beck K, D’Ambrosio E, Donocik J, Gobjila C, Jauhar S, et al. Antipsychotic plasma levels in the assessment of poor treatment response in schizophrenia. Acta Psychiatr Scand. 2018;137(1):39-46. doi:10.1111/acps.12825

32.      Hiemke C, Bergemann N, Clement H, Conca A, Deckert J, Domschke K, et al. Consensus guidelines for therapeutic drug monitoring in neuropsychopharmacology: Update 2017. Pharmacopsychiatry 2018;51(01/02):9-62. doi:10.1055/s-0043-116492

33.      Schoretsanitis G, Kane JM, Correll CU, Marder SR, Citrome L, Newcomer JW, et al. Blood levels to optimize antipsychotic treatment in clinical practice. J Clin Psychiatry. 2020;81(3). doi:10.4088/JCP.19cs13169

34.      Samara MT, Dold M, Gianatsi M, Nikolakopoulou A, Helfer B, Salanti G, et al. Efficacy, acceptability, and tolerability of antipsychotics in treatment-resistant schizophrenia. JAMA Psychiatry. 2016;73(3):199. doi:10.1001/jamapsychiatry.2015.2955

35.      Lindenmayer J-P, Citrome L, Khan A, Kaushik S, Kaushik S. A randomized, double-blind, parallel-group, fixed-dose, clinical trial of quetiapine at 600 versus 1200 mg/d for patients with treatment-resistant schizophrenia or schizoaffective disorder. J Clin Psychopharmacol. 2011;31(2):160-168. doi:10.1097/JCP.0b013e31820f4fe0

36.      Meltzer HY, Bobo WV, Roy A, Jayathilake K, Chen Y, Ertugrul A, et al. A Randomized, double-blind comparison of clozapine and high-dose olanzapine in treatment-resistant patients with schizophrenia. J Clin Psychiatry. 2008;69(2):274-285. doi:10.4088/JCP.v69n0214

37.      Meltzer HY, Lindenmayer JP, Kwentus J, Share DB, Johnson R, Jayathilake K. A six month randomized controlled trial of long acting injectable risperidone 50 and 100 mg in treatment resistant schizophrenia. Schizophr Res. 2014;154(3):14-22. doi:10.1016/j.schres.2014.02.015

38.      Siskind D, McCartney L, Goldschlager R, Kisely S. Clozapine vs. first- and second-generation antipsychotics in treatment-refractory schizophrenia: systematic review and meta-analysis. Br J Psychiatry. 2016;209(5):385-392. doi:10.1192/bjp.bp.115.177261

39.      Kawai N, Yamakawa Y, Baba A, Nemoto K, Tachikawa H, Hori T, et al. High-dose of multiple antipsychotics and cognitive function in schizophrenia: The effect of dose-reduction. Prog Neuro-Psychopharmacology Biol Psychiatry. 2006;30(6):1009-1014. doi:https://doi.org/10.1016/j.pnpbp.2006.03.013

40.      Simon V, van Winkel R, De Hert M. Are weight gain and metabolic side effects of atypical antipsychotics dose dependent? J Clin Psychiatry. 2009;70(7):1041-1050. doi:10.4088/JCP.08r04392

41.      Álamo C, López-Muñoz F. The pharmacological role and clinical applications of antipsychotics’ active metabolites: paliperidone versus risperidone. Clin Exp Pharmacol. 2013;3:1-12. doi:10.4172/2161-1459.1000117

42.      De Hert M, Correll CU, Cohen D. Do antipsychotic medications reduce or increase mortality in schizophrenia? A critical appraisal of the FIN-11 study. Schizophr Res. 2010;117(1):68-74. doi:10.1016/j.schres.2009.12.029

43.      Taipale H, Mittendorfer-Rutz E, Alexanderson K, Majak M, Mehtälä J, Hoti F, et al. Antipsychotics and mortality in a nationwide cohort of 29,823 patients with schizophrenia. Schizophr Res. 2018;197(2018):274-280. doi:10.1016/j.schres.2017.12.010

44.      Kishi T, Matsunaga S, Iwata N. Mortality risk associated with long-acting injectable antipsychotics: A systematic review and meta-analyses of randomized controlled trials. Schizophr Bull. 2016;42(6):1438-1445. doi:10.1093/schbul/sbw043

45.      Osser DN. A Systematic approach to pharmacotherapy in patients with neuroleptic-resistant psychoses. Psychiatr Serv. 1989;40(9):921-927. doi:10.1176/ps.40.9.921

46.      Osser DN. Prolactin monitoring in first-episode psychotic patients. Schizophr Res. 2017;189:2-3. doi:10.1016/j.schres.2017.10.001