Current Data on and Clinical Insights into the Treatment of First Episode Nonaffective Psychosis: A Comprehensive Review

The first months of antipsychotic treatment represent the critical period for changes in weight [61, 62]. Longitudinal studies in drug-naïve subjects have shown that the percentage of obese (body mass index (BMI) >30) subjects increased from 5% to 32% and the percentage of overweight rose up to 65% at 3 years after the initiation of treatment with olanzapine, risperidone, or haloperidol [62]. Marked differences in likely weight gain exist among the different SGAs. Of clinical relevance is the fact that several reviews and studies have shown that olanzapine and clozapine are associated with a greater risk of weight gain and higher percentage of extreme weight gain (>7%) and aripiprazole and ziprasidone have a low risk of weight gain [52, 60, 63]. In a recent study in drug-naïve individuals with a FEP, aripiprazole and quetiapine showed a significant increase in weight compared to ziprasidone [54]. Studies on drug-naïve subjects are in agreement with this statement, in particular during the first months of treatment [45, 64, 65]. The EUFEST study revealed that ziprasidone has a lesser effect on weight gain than quetiapine and olanzapine [31, 66]. Low doses of risperidone and aripiprazole show a similar pattern of weight gain during early phases of treatment in people with a FEP [33, 54].

Accordingly, aripiprazole and ziprasidone have shown lower liability for inducing weight gain and metabolic adverse reactions than olanzapine and risperidone in long-term schizophrenia populations [64, 67‐69]. Furthermore, several studies have shown that switching from treatments with a high risk of weight gain to ziprasidone [70, 71] or to aripiprazole [72] is an effective strategy to reduce weight and improve metabolic outcomes in people with stable chronic schizophrenia or schizoaffective disorder. Finally, short-term, double-blind clinical trials comparing ziprasidone with placebo parameters [73, 74] or aripiprazole with placebo [75] in people with chronic schizophrenia or schizoaffective disorder have not found significant differences in their effects on metabolic variables.

The severity of weight increase, and therefore fatty change, is correlated with hepatic and cardiovascular toxicity. Thus, obesity is a risk factor for nonalcoholic fatty liver disease (NAFLD), which is the most common cause of liver disease in Western countries [76]. Overall, it has been described that 50% of people with a BMI >30 may suffer NAFLD. In a recent study, it has been demonstrated that over 25% of individuals with a FEP will develop liver steatosis after 3 years of treatment (Morlan et al. 2015 submitted, personal communication). Because of the link between cardiometabolic risk, NAFLD, and antipsychotic medication, early detection in clinical practice of NAFLD development, based on non-invasive techniques, could be worthy of consideration, as these subjects should be considered for liver disease treatment and for monitoring and potential treatment of underlying cardiovascular risk factors [77]. It has been observed that individuals treated with olanzapine who gained a substantial amount of weight manifested significantly lower heart rate variability (a marker of high cardiovascular risk) [78].

In addition to physical disturbances, weight stigmatization and discrimination lead to important psychological stress that increases the risk for depression, low self-esteem, and body dissatisfaction in people with schizophrenia [79]. In the long term, weight gain may lead to poorer social and functional outcomes [80] and weight loss is linked to an improvement in social functioning [81] in people with schizophrenia.

Behavioral and counseling interventions including smoking cessation, dietary measures, and exercise might be partly effective in reducing weight gain and metabolic disturbances in early stages of treatment [82‐84]. Nonetheless, the benefits of these lifestyle interventions do not seem to persist across time [85]. The preferential use of or switching to medications with a lower risk of weight and/or metabolic abnormalities is recommended [86].

The risk of glucose intolerance and diabetes associated with antipsychotic treatments seems to be increased in younger subjects [87]. After 3 years of antipsychotic treatment, insulin plasma levels and the HOMA index tend to increase in FEP individuals [62]. Olanzapine, clozapine, and quetiapine have been associated with more prevalent hyperglycemia independently of body mass in both drug-naïve [88] and chronic schizophrenia individuals [89, 90]. Higher levels of insulin and insulin resistance were associated with olanzapine treatment in patients with FEP or schizophrenia and in healthy normal-weight men [31, 91, 92]. Quetiapine treatment has also been shown to increase the risk of triglycerides to HDL-C ratio (a marker of insulin resistance) in FEP [50]. A critical factor to bear in mind by clinicians is the fact that type 2 diabetes might have an onset a median of 4 years after first antipsychotic treatment, although subjects on olanzapine had a shorter time to diabetes onset [88]. It has also been reported that olanzapine discontinuers did not have an increased risk of diabetes [88]. Higher antipsychotic dosages and male sex seem to be related to an increased risk for diabetes. Due to the close association between metabolic effects and the greater risk of diabetes, pre-diabetes or type 2 diabetes should be prevented or postponed by choosing as a first choice an antipsychotic with a lower cardiometabolic burden.

After 3 years of antipsychotic treatment, a high proportion of FEP individuals may have been exposed to high plasma total triglyceride concentrations (up to 23%), low levels of HDL-C (23%), and a high total cholesterol/HDL-C ratio (29%) [62]. Olanzapine and quetiapine have shown the greatest effect on lipid status, whereas aripiprazole and ziprasidone have produced the lowest effects on lipid levels in both FEP and chronic schizophrenia populations [32, 93]. Our group has described that after 12 weeks of treatment ziprasidone has a neutral effect on fasting plasma lipids and glucose [54]. Importantly, increases in dyslipidemia and triglycerides have been shown to be meaningfully associated with olanzapine treatment independent of weight gain in both FEP and chronic schizophrenia [94, 95]. In a recent study, Robinson and colleagues (2015) observed that cholesterol and LDL-C levels, but not triglycerides and HDL-C, were significantly less affected in individuals on aripiprazole compared to risperidone-treated FEP patients [33].

Hyperprolactinemia is considered to be a disturbing adverse effect of antipsychotic medication, although recent investigations have described that it might be also a pre-existing or stress-related condition, since it has been described in antipsychotic-naïve people with FEP [96‐99] and in subjects with an at-risk mental state [96]. Antipsychotic medications differ in their propensity to induce hyperprolactinemia (for a review see Peuskens et al. [100]). Most FGAs have been associated with a pronounced elevation of prolactin levels in the treatment of FEP [30, 101], but some investigations in FEP have described no significant prolactin changes at long term associated with haloperidol treatment [31, 98]. Some of the SGAs are considered prolactin-sparing antipsychotics and quetiapine, ziprasidone, olanzapine, clozapine, and aripiprazole are less likely to induce sustained hyperprolactinemia [31, 39, 54, 98, 102], but others have been associated with significant increases in plasma prolactin levels, and thus are considered to be prolactin-elevating antipsychotics – risperidone, amisulpride, and paliperidone [31, 101]. Nonetheless, clinicians should be fully aware that 40% of the subjects on prolactin-sparing antipsychotics might exhibit hyperprolactinemia. Plasma prolactin level elevations with antipsychotic drugs are generally dose dependent [100].

Many patients with elevated prolactin levels are asymptomatic, and others may not report symptoms unless specifically questioned [103]. Sustained elevation of prolactin may cause amenorrhea, galactorrhea, hirsutism, gynecomastia, impotence, loss of libido, and infertility in the short term. Long term hypogonadism due to hyperprolactinemia may lead to osteoporosis and hip fracture [100]. The expected utility of switching to or adding aripiprazole to rapidly decrease elevated prolactin levels induced by prolactin-elevating antipsychotics has been described in samples of patients with chronic schizophrenia [104, 105].

Antipsychotic-associated hyperprolactinemia is the most common cause of elevations in prolactin levels in individuals with schizophrenia, but additional appropriate investigations may also be warranted for those individuals with persistently elevated prolactin levels [106]. Clinicians should consider reducing the antipsychotic dosage or switching to an antipsychotic drug with a lower potential to elevate prolactin in cases of hyperprolactinemia.

Sedation is a common and unpleasant adverse effect of both FGAs and SGAs [18], with young subjects seemingly more affected by somnolence and hypersomnia than older adults [107, 108]. Given that sedation-related side effects may particularly cause discomfort and daytime dysfunction, this effect may lead to patients becoming non-adherent to treatment in patients with schizohrenia [109]. Some clinicians may regard sedation as a helpful effect during an acute episode for addressing symptoms such as insomnia and agitation and may use antipsychotics with sedative effects, antipsychotic combinations or higher doses in order to obtain sedation. However, in a phase-specific treatment, sedation may no longer be needed and should be closely monitored and prevented, including the reduction of antipsychotic doses, avoiding polypharmacy, or switching the antipsychotic [110]. Leutch et al. showed evidence that clozapine and chlorpromazine are associated with higher sedation, and amisulpride and paliperidone certainly produce less sedation in the treatment of individuals with schizophrenia [111].

Young people suffering from a first episode of psychosis seem to be highly responsive to low doses of antipsychotics and more susceptible to extrapyramidal side effects and to acute weight gain [112, 113]. In the last few decades, chlorpromazine doses in people with schizophrenia and other psychoses have been gradually reduced and this has probably led to less extrapyramidal side effects [114]. A clear limitation when establishing the lower effective antipsychotic dose in the FEP population is that available dose-equivaleny methods for antipsychotics are not applicable to FEP [115]. Additionally, people with a FEP are more likely to respond to lower doses of antipsychotic medications and treatment should be started at the lower half of the recommended dosage range for multiple episodes (300–500 mg chlorpromazine equivalents per day) [116]. There is a clear recommendation for using lower doses of antipsychotic medications in cases of FEP [27, 116, 117]. Previous studies assessing the efficacy of antipsychotics in FEP have evaluated the efficacy of lower doses. The mean modal doses of risperidone have ranged from 2.4 to 4 mg/day [41, 101, 118, 119], and of olanzapine from 9.1 to 12.6 mg/day [31, 41, 120]. Minimum effective doses of 2 mg of haloperidol [121] or lower and 2 mg of risperidone [120] have been previously reported in FEP. As an example, in our cohort, FEP was effectively treated with initial modal doses at 6 weeks of 5 mg of haloperidol (range 3–9 mg/day), 4 mg of risperidone (3–6 mg/day), 15 mg of olanzapine (5–20 mg/day) [49], 15 mg of aripiprazole (5–20 mg/day), 80 mg of ziprasidone (40–160 mg/day), or 300 mg of quetiapine (100–600 mg/day) [36]. In a recent investigation Robinson et al. used mean daily doses of 14.8 mg for aripiprazole and 3.2 mg for risperidone [33]. Even with a “start low, go slow” model, given that people with a FEP require low doses, they may be treated with effective doses from the first week and the antipsychotic rapid titration should allow adjusting doses to avoid side effects.

Although doses may be reasonably increased within the therapeutic range if a clinical response is not obtained, an increase in the antipsychotic dose is less likely to be successful in FEP, so using high antipsychotic doses is discouraged [35].

It has been suggested that the long-term use of antipsychotics may lead to a dopaminergic supersensitivity state, caused by an upregulation of the dopamine receptors [122] that would predispose patients with psychosis who receive longer and higher doses to relapse. However, studies that have tried to test this theory have failed to prove it, as they did not find differences according to medication type withdrawn or tapered [123].

According to international treatment guidelines, there is uncertainty about the recommended time clinicians should wait before switching medication, suggesting an initial trial of “appropriate doses” of antipsychotics from 2 to 6 weeks according to current guidelines [124, 125]. The risk-benefit assessment clearly favors switching antipsychotics when a given medication is clearly ineffective. Nevertheless, clinicians, before making the decision to switch, should ensure that treatment with the pre-switch antipsychotic was optimized in terms of adherence and dose [126]. In a recent meta-analysis, Samara et al. reported that people with schizophrenia who did not even minimally improve at week 2 of treatment with sufficiently high doses are unlikely to respond later and therefore may benefit from a change or augmentation of treatment. Ninety percent of subjects who do not show a minimal response (20% reduction in the PANSS/BPRS score) at the 2-week assessment will not show clinical improvement at week 6 [127]. Although the application of these results seems to be more appropriate for people who are not in their first episode of schizophrenia, they emphasize the need for a rapid titration model in order to achieve an adequate dose during the first days of treatment. Additionally, clinicians dealing with FEP individuals should be fully aware of the necessity of switching or implementing augmentation treatment strategies early in the course of the antipsychotic treatment to avoid unnecessary long-term exposure to an antipsychotic that is very unlikely to be helpful. Lack of clinical efficacy is among the reasons that might prompt people with an FEP to treatment discontinuation [19]. Strategies for switching have been previously poorly studied in an attempt to make an accurate recommendation, but the future results of two ongoing trials, SWITCH and OPTiMiSE, should help decision making and provide a basis for clinical guidelines. In a previous finding, Agid et al. [128] described only a 20% responsiveness among subjects who did not respond to the first-line antipsychotic drug, whereas the efficacy rate of clozapine in these nonresponsive subjects was 75%; therefore early after a lack of response, a second antipsychotic trial and then the use of clozapine may be indicated when other issues such as adherence or co-morbidity have been addressed [27, 125].

Likely problems in switching antipsychotics in the treatment of schizophrenia such as the risk of discontinuation reactions and the re-emergence of psychotic symptoms have been traditionally described [129, 130]. Thus, overall patients should be monitored carefully throughout the switch procedure for any discontinuation effects or new side effects [131, 132]. More frequent contacts might be needed during the switching process in order to reassure the patient and to monitor for side effects and discontinuation problems [133, 134].

As with individuals with several episodes, using antipsychotic polypharmacy is discouraged due to a lack of proven efficacy and a higher risk of side effects [125, 135]. Importantly, co-morbid symptoms such as depression, anxiety, suicide risk, and substance misuse must be correctly diagnosed and treated in order to avoid early refractoriness, lack of remission, and relapse [27, 125]. Subjects who not respond to the first antipsychotic trial after a few weeks have a low response rate thereafter. Addressing co-morbidity must include not only choosing the appropriate antipsychotic medication but also offering adequate psychosocial interventions such as cognitive behavior therapy (CBT) or treating associated problems such as substance abuse [136, 137]. CBT has been reported as a useful treatment for schizophrenia in several reviews [138‐141], particularly in the treatment of persistent positive symptoms as part of a combined therapy. This has led to the inclusion of CBT into services and recommendations of treatment guidelines for people suffering psychotic disorders, including a FEP [27, 125]. Moreover, CBT seems to be helpful in improving other symptoms such as negative symptoms, functioning, mood, and social anxiety [142‐145]. However, Jauhar et al. described in a meta-analysis of CBT effectiveness for symptoms of schizophrenia that pooled effect sizes were in the “small” range for all the classes of symptoms considered (pooled effect size of −0.33 for overall symptoms; −0.25 for positive symptoms; negative sign favors CBT) [146]. Previous recent meta-analyses by Wykes et al. had described higher levels of effectiveness that were in any case modest (i.e., pooled effect size of −0.37 for positive symptoms, pooled effect size of −0.44 for negative symptoms) [147]. With regard to substance misuse, integrated CBT and motivational interviewing have been used with the aim of reducing the use of substances and reduce subsequent relapses in a FEP; however, results of these interventions are not clearly consistent [137, 148‐150]. Although there are some ongoing trials evaluating the specific efficacy of CBT and motivational interviewing in reducing the use of some substances such as cannabis [151], it is possible that the benefits of these psychosocial interventions are best reflected in their effects on symptomatic outcomes rather than in a reduction of substance use or relapse prevention, and therefore the appropriate strategy may be to address associated problems rather than the substance use per se [137]. CBT for psychosis (CBTp) cannot be considered as a global brand because not all approaches are the same. It is now highly recommended that choosing the most appropriate CBTp should take into account patient-defined goals and critical targets and the types of individuals who can benefit from therapy [152].

In clinical practice, due to the lack of insight into the prophylactic effects of drugs, emergence of disturbing side effects, and concerns over the long-term harm associated with antipsychotics, it is unlikely that patients will accept an indefinite treatment after a single episode of psychosis [20]. However, stopping antipsychotic medication has been repeatedly demonstrated as the biggest predictor of relapse after a first episode [153‐155]. Wunderink et al. [156], exploring people that remitted after a FEP, observed that relapse rates were two times higher in individuals who gradually tapered or discontinued medication and only 20% of subjects could be successfully discontinued. Strikingly, those patients with an earlier reduction or discontinuation treatment strategy appear to have long-term functional gains compared with individuals who maintained treatment [157, 158]. In a recent investigation, Mayoral-van Son et al. [20] described a high risk of symptom recurrence after antipsychotic treatment discontinuation in individuals who had accomplished a functional recovery after a single psychotic episode. Moreover, those relapsed individuals had a greater severity of symptoms and lower functional status after 3 years. The profile of individuals who would benefit from treatment discontinuation, and the optimal duration of prophylactic antipsychotic medication remain unclear. The lowest effective-dose regimen is highly recommended [20, 154] and a trial off antipsychotic medication is not to be recommended [159]. Clinicians should provide accurate information to people with a FEP (and relatives) about the fact that even if they have been symptom free and functionally recovered on antipsychotic treatment, the risk of relapse is still likely to be high if antipsychotics are discontinued [160]. Despite these clear statements, numerous individuals may still be disposed to discontinue medication after recovering from their single episode of psychosis and, therefore, a planned medication withdrawal strategy with a systematic follow-up should be established in first-episode programs to prevent unrestrained treatment discontinuation.

Independent of the initial high response rate after a FEP, during follow-up of the illness, the relapse rates are high, being up to 83% 5 years after the FEP [153]. Persistent substance use disorder, carer criticism, poorer premorbid adjustment, and, particularly, medication nonadherence (a fourfold risk) may significantly increase the risk for relapse in FEP [161]. In a previous report analyzing a subsample of 140 first episode individuals, we observed a similar relapse rate of 65% at 3 years after the first break of the illness, being adherence to antipsychotic treatment the only significant predictor for relapse [155]. Therefore, in order to prevent relapse, the best way would be to improve adherence and to reduce other avoidable risk factors such as substance use or carer criticism. A basic strategy that may help clinicians improve their patients’ adherence includes improving prescriber-patient communication. The clinician should listen the patient, understand his or her perspective, beliefs and concerns about the illness and treatment, and should provide information regarding potential effectiveness, side effects, and formulations in order to get the patient involved in decisions about their medications [162]. The use of psychosocial treatments, including psychoeducational interventions, CBT, and individual and family interventions, may specifically improve adherence and must be offered in combination with antipsychotic treatment in order to prevent relapses in patients with a FEP [163]. Finally, after the evaluation of adherence and associated risks and the provision of adequate medication, psychoeducation and long-acting injectable (LAI) antipsychotics should be offered in order to improve outcome in selected subjects [164, 165]. The use of LAI antipsychotics has been traditionally restricted to people who are noncompliant and who have a history of multiple relapses, and FEP is rarely treated with depot medications [166]. Early introduction of LAI therapy in FEP has been associated with a short-term adherence benefit [167]. However, subsequent follow-up of the same cohort did not reveal differences between patients treated with LAI therapy and patients with oral antipsychotics in time to initial non-adherence at 104 weeks [168]. Although more well designed, randomized, controlled clinical trials using SGA LAIs exploring effectiveness and likely long-term side effects are warranted, LAIs should be given due consideration in early phases treatment of based on their benefits with regard to adherence [169]. In a recent review of the literature on the use of LAIs in first-episode and early schizophrenia, Taylor and Ng concluded that LAIs are useful in reducing symptom severity and the risk of relapse, particularly if they voluntary accept LAI therapy [170]. Ironically, clinicians are often reluctant to offer LAIs in early phases and tend to overestimate potential objections by patients [171].

After obtaining a clinical response, accomplishing symptomatic and functional remission is one of the major objectives of treating individuals with a FEP, with a final objective of functional recovery. In 2005, the Remission in Schizophrenia Working Group published the most widely accepted criteria to define remission in schizophrenia [172]. This concept includes the remission of positive psychotic, disorganized, and negative symptoms to a severity of mild or less, but also a minimum duration time for this remission of 6 months, and has been used in FEP. Rates of remission vary widely in different studies and most of them do not provide duration criteria [173]. As an example, after a 1-year follow-up, only 31% of the initial subjects enrolled in the PAFIP program achieved remission [174], in particular due to persistence of negative symptoms or to relapse in the last 6 months, which seem to be related to duration of untreated psychosis and premorbid functioning. Remission is closely related to functional recovery, although symptomatic remission rates tend to exceed functional recovery rates [173]. The best ways to achieve remission after improving initial symptoms seem to be preventing relapses and also detecting and treating secondary negative symptoms as early as possible. Additionally, vocational interventions have been recognized to be beneficial to achieve better functional outcomes [175].

One of the main objectives of FEP programs and early intervention services is to engage with both people with psychosis and their family and to facilitate support. However, until recently, most of the effort has gone into studying, recognizing, and reducing expressed emotion (EE) and its components – critical comments (CC) and emotional over-involvement (EOI) – due to its potential prediction of relapse and symptom exacerbation in schizophrenia. Given that a large majority of people during their FEP live at home with their families, these informal caregivers are affected by a stress-appraisal-coping framework [176]. This would mean that the extent to which the stressors are related to the relative’s or caregiver’s negative health is related to his/her appraisal (or subjective evaluation) of the stressor and his/her ability to cope with them. Apart from family interventions, basically aimed at preventing or reducing relapses, psychoeducation and support groups have been developed to improve the experience of caring for people with mental illness, and particularly may be recommended to families and caregivers of people with FEP and severe mental disorders [177]. The role of FEP programs and early intervention services is to detect the needs of the families and caregivers and give sustained appropriate communication and support, in order to improve their experience of caring and reduce associated stress.

Although the antidopaminergic mechanism (need for D2R blockade) is still essential in the development of new antipsychotics, the investigation of new atypical antipsychotics has focused on either compounds with a broader spectrum of clinical benefits through dual dopaminergic and serotoninergic properties – “serotonin-dopamine activity modulators”– (e.g., partial agonism of D2R with enhanced affinity for specific serotonin receptor analog of aripiprazole) or exploiting less examined mechanisms (e.g., cholinergic, histaminergic, GABAergic, modulation of glutamate); for a review see Zajdel et al. [178]. Brexpiprazole, with structure, properties and uses that are most similar to those of aripiprazole, has recently exhibited good tolerability, with a lower incidence of akathisia than aripiprazole [179]. Initial animal studies with cariprazine, a potent D3/D2 receptor partial agonist with favored binding to D3 receptors, seem to point out it could be a better treatment option for patients with persistent and predominant negative symptoms [180]. The long half-life of cariprazine’s principal active metabolites and the steady-state levels of cariprazine and metabolites within 4 weeks are pharmacological characteristics that may minimize the impact of missing doses on plasma levels [181]. Significant improvement in hostility has been also observed in cariprazine- versus placebo-treated patients [182]. Iloperidone shares a high affinity for both D2 and 5HT2A receptors, with a unique receptor-binding profile that includes a very strong affinity for the noradrenergic alpha 1 (NEa1) receptor [183]. It is a well-tolerated antipsychotic with a very low propensity to cause either akathisia or antipsychotic-induced extrapyramidal symptoms [184]. Lurasidone exerts a potent antagonism toward the dopamine D2, and the 5-hydroxytryptamine (5-HT, serotonin) receptors: 5-HT2A and 5-HT7 [185]. Apparently lurasidone is less efficacious than clozapine, risperidone, paliperidone, olanzapine, and amisulpride, with a lower risk of sedation and body mass increase [111, 186]. However, subjects treated with lurasidone exhibited higher rates of akathisia, parkinsonism, and hyperprolactinemia compared to several other atypical antipsychotics. To the best of our knowledge none of the above new SGAs have been studied in first-episode cohorts to date.