Low Dose Loxapine: Neuromotor Side Effects and Tolerability in Autism Spectrum Disorders

Introduction

Autism spectrum disorders (ASD) are neuropsychiatric disorders that impair the ability to communicate and interact with others, and are lifelong. ASD ranges in severity and is classified based on severity of impairments in social interactions and communication skills, and repetitive behaviors (American Psychiatric Association 2013). Individuals with ASD are treated primarily with educational supports, speech therapy, physical therapy, occupational therapy, and behavioral interventions. Psychotropic medications have so far only been identified to treat comorbidities of ASD such as attention-deficit/hyperactivity disorder, obsessive compulsive disorder, or bipolar disorders (Fletcher et al. 2007), and not core symptoms.

Few drugs have received adequate study for treatment of serious challenging behaviors associated with ASD. Risperidone and aripiprazole are the only atypical antipsychotics shown to decrease such behaviors, including aggression, irritability, and self-injury, in large-scale, double-blind placebo-controlled studies in subjects with autistic disorder (McCracken et al. 2002; Marcus et al. 2011; Benvenuto et al. 2013). However, atypical antipsychotics are associated with marked appetite increase and lack of satiation (Allison et al. 1999), possibly because of the drug action on serotonin 5-HT2 receptors in the hypothalamus, which plays a key role in feeding behavior and satiation. Numerous studies have linked atypical antipsychotics with weight gain, dyslipidemia, insulin resistance, and type II diabetes (Hellings et al. 2001; Newcomer and Haupt 2006; Hellings et al. 2010, 2011). These effects increase risk for impaired mobility, cardiovascular disease, arthritis, and various forms of carcinoma, enormously increasing public health costs and reducing survival rates and quality of life, for this population with preexisting increased risks.

Therefore, antipsychotics that have reduced risk for both metabolic disturbances and neuromotor side effects are urgently needed. The multisite Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study in adults with schizophrenia demonstrated that classical antipsychotics remain effective, cost much less, and have different adverse event profiles than novel antipsychotics (Lieberman and Stroup 2011). Prospective studies of classical antipsychotics in ASD mostly focused on haloperidol, which has significant side effects, including extrapyramidal side effects (EPS), tardive dyskinesia (TD), withdrawal dystonias and neuroleptic malignant syndrome (NMS) (Campbell et al. 1997; Leskovec et al. 2008; Posey et al. 2008).

Following the development of atypical or second-generation antipsychotics (SGAs) in the 1990s, these drugs have been prescribed more commonly than classic antipsychotics. This relates to marketing efforts as well as the tendency of classic antipsychotics to cause EPS and other neuromotor side effects, including akathisia, TD, and NMS in usual therapeutic doses. Such neuromotor symptoms are more likely to occur in individuals with developmental disabilities and mobility issues associated with neuromuscular disorders such as cerebral palsy, Prader–Willi syndrome (Reus et al. 2012), and Rett syndrome. At the same time, individuals with developmental disabilities that include language impairments are less likely to be able to report such side effects and, therefore, great clinician vigilance is required.

EPS include muscle stiffness or rigidity, shuffling gait, mask-like face, and resting tremor, which not only are dose and metabolism related, but also increased by underlying neuromotor defects. Akathisia or motor restlessness is another dose-related side effect of classic and atypical antipsychotics, and occurs in susceptible individuals especially if the starting antipsychotic dose is high. TD may develop after years of treatment with classic or atypical antipsychotics, and comprises worm-like tongue movements initially, followed later by whole tongue-thrusting and lip-chewing movements, grimacing, and facial movements, and, eventually, also whole-body jerking (Jibson 2014). TD may be irreversible even with taper and discontinuation of the antipsychotic, and occurs in 5.5–32.4% of those taking classic antipsychotics, according to a sample set study (Correll and Schenk 2008).

Concomitant use of selective serotonin reuptake inhibitors (SSRIs) including paroxetine, fluoxetine, or sertraline also increases the effective antipsychotic dose significantly, increasing the likelihood of neuromotor impairments and metabolic side effects (Preskorn 1997).

Loxapine in low doses of 5–15 mg/day shows promise for use in ASD, to treat irritability and aggression, as well as for psychosis and mood stabilization. This is based on clinical experience using lowest possible doses for efficacy and safety. Although labeled as a medium potency classic antipsychotic, low dose loxapine has both typical and atypical antipsychotic properties because of predominant blockade of dopamine D2 receptors as well as less potent serotonin 5-HT2 and acetylcholine receptor antagonism (Ereshefsky 1999; Li et al. 2003; Stahl 2008). In a published case report of a 10-year-old African American girl with autistic disorder, Reinblatt et al. (2006) suggest that loxapine may be a suitable option to treat refractory irritability, aggression, and destructive behavior in patients with ASD, particularly those with risk factors for metabolic syndrome. Our group recently reported on the success of adding loxapine 5–15 mg daily in a 12 week prospective open add-on trial for adolescents and adults with ASD and irritability. All 14 subjects showed the primary outcome measure of ≤2 (Much Improved or Very Much Improved) on the Clinical Global Impressions Improvement (CGI-I) subscale, and mean change on the Aberrant Behavior Checklist-Irritability (ABC-I) subscale was −31% by week 12 (Hellings et al., 2015). Furthermore, neuromotor and other side effect rates were low in this study, with only one subject requiring benztropine to treat EPS. Importantly, loxapine was weight neutral.

Because of its unique profile of potent D2 and lesser 5-HT2 receptor blockade, low dose loxapine merits further study. Its reduced propensity for common side effects occurring with both typical and atypical antipsychotics, along with its promising efficacy in improving treatment-resistant irritability and aggression, make loxapine an important atypical antipsychotic-sparing option (Ereshefsky 1999; Stahl 2008; Hellings et al., 2015). In the present study, we cross-sectionally examined for neuromotor side effects following short- and long-term use of loxapine in children, adolescents, and adults with ASD in the outpatient setting. Specifically, we documented any occurrence of EPS, akathisia, NMS, and TD during loxapine treatment. We hypothesized based on our clinical experience using low dose loxapine for >15 years, even together with atypical antipsychotics, that EPS and akathisia would be mild, and TD more rare than expected with full dose classical antipsychotics, even after chronic use. Rates of TD with classic antipsychotics are 5.5–32.4% (Correll and Schenk 2008). No cases described in this study have already been published.

Methods

Approval for the study was obtained from the University of Kansas Medical Center Institutional Review Board. Consecutive outpatients with ASD attending the neuropsychiatry clinic at the University of Kansas Medical Center who received loxapine treatment were invited to participate in the study. Informed, written consent was obtained for each subject, as well as assent when possible.

Study methods comprised two components: A prospective one that was a cross-sectional neuromotor examination in the clinic for side effects, and, thereafter, a retrospective chart review for each subject examined, to extract information pertinent to loxapine treatment. Inclusion criteria were: 1) Males and females 5–65 years of age; 2) Diagnostic and Statistical Manual of Mental Disorders, 4th ed., Text Revision (DSM-IV-TR) diagnosis of Autistic Disorder, Pervasive Developmental Disorder- Not Otherwise Specified (PDD-NOS), or Asperger Disorder; and 3) treatment with loxapine for ≥6 months. Exclusion criteria were: 1.) poor medication compliance; 2.) multiple rescheduling of appointments, impeding treatment; and 3) presence of neurodegenerative disorders. All diagnoses were made in accordance with Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV) or DSM-IV-TR criteria (American Psychiatric Association 1994, 2000), and for bipolar disorder also using the approach outlined in the review by Hellings (1999).

Data were extracted as part of the retrospective chart review and checked independently by two separate investigators as follows: Age at loxapine treatment initiation, race, gender, ASD subtype, dosage and duration of loxapine treatment, and concomitant medications. Any reduction in loxapine dosage was also noted, along with reasons for reduction. For measuring behavioral outcomes, Clinical Global Impressions- Severity (CGI-S) (Guy, 1976), and CGI-I ratings, as well as side effects, had been rated by the same rater (J.A.H.) at each clinic visit, and were extracted for the visit when loxapine was started and for the visit when the study examination was performed. Treatment response was defined as ≤2 for the CGI-I score at the time of neurological examination. Any missing ratings were determined by J.H. using the clinic progress note.

Neuromotor side effects of loxapine were measured with standardized rating scales. TD was measured by the Dyskinesia Identification System Condensed User Scale (DISCUS) (Sprague et al. 1989), which takes into account involuntary movements such as lip-smacking, facial grimacing, lateral jaw movements, and choreoathetoid movements of the tongue, extremities, or trunk. EPS were measured using the Neuroleptic Side Effects Checklist (Gualtieri 1984), which documents EPS and acute dystonias. Akathisia was examined using the Barnes Akathisia Rating Scale (Barnes 1989, 2003), which distinguishes between subjective feelings of restlessness and objective findings of fidgeting, pacing, or inability to sit still (noting that baseline pacing and motor restlessness is a confound to accurate measurement).

We performed a descriptive analysis of the data.

Results

Thirty-four subjects completed the examination. As shown in Table 1, 25 were males and 9 were females. Mean age was 23.4 years at start of loxapine treatment (range 8–52 years). Mean loxapine dose was 8.9 mg daily (range 5–30 mg daily). Fifteen subjects (44.1%) received loxapine for at least 5 years. Mean duration of loxapine treatment was 4.2 years (range 0.8–13 years). Fourteen subjects (41.2%) also received atypical antipsychotics. Loxapine dose was decreased in only one subject, from 15 mg daily down to 10 mg daily because of observed behavioral worsening after attempted increase.

Seventeen subjects met DSM-IV-TR criteria for autistic disorder, 15 for PDD-NOS, and 2 for Asperger Disorder. All had comorbid psychiatric diagnoses. Bipolar mood disorder was also diagnosed in 14 of 34 subjects, schizoaffective disorder in 3, attention-deficit/hyperactivity disorder in 19, obsessive compulsive disorder in 18, and intermittent explosive disorder in 7. Tourette disorder, compulsive polydipsia, and separation anxiety disorder were each diagnosed in one subject, as was posttraumatic stress disorder. Thirteen subjects were aggressive on presentation, and 15 had manifested self-injury.

Tremor was noted on the Neuroleptic Side Effects Scale (unpublished scale available from author) in seven subjects (20.5%). Five subjects (14.7%) received benztropine, dosed at 1–2 mg daily for EPS management, and subject 32 also received propranolol for tremor that was worsened by concomitant lithium and valproic acid. Concomitant medications that likely contributed to tremor by increasing effective dose through cytochrome P450-2D6 (CYP2D6) inhibition were: Paroxetine (one subject) and sertraline (one subject). Other medications worsening tremor included dextroamphetamine and atomoxetine used together in one subject, and valproic acid in two subjects. Three subjects with tremor received loxapine and an atypical antipsychotic, whereas four received loxapine as the only antipsychotic. Eleven subjects who received loxapine and an atypical antipsychotic did not manifest tremor. As observed using DISCUS, three subjects (8.8%), 18, 22, and 23, manifested tics prior to loxapine treatment. Final DISCUS TD item scores for these subjects were 0 (no TD movements), 2 (chewing lips), and 1 (lip-smacking), respectively, thus showing improvements in their baseline tics.

One of 34 subjects (2.9%) showed TD, and his was an exceptional case. This was subject 26, an adult Caucasian male with Prader–Willi syndrome, who manifested TD-type mouth movements: His DISCUS score was 5, comprising chewing/lip smacking rated at 3 and lower lip thrusting rated at 2. His loxapine dose of 5 mg daily was likely much elevated by concomitant treatment with paroxetine 40 mg daily, an SSRI that is a potent CYP2D6 liver enzyme inhibitor (Preskorn 1997). He did not receive any concomitant atypical antipsychotic treatment.

Objective and subjective akathisia was measured using the Barnes Akathisia Rating Scale. Objective scores were all zero except in six subjects (17.6%). Out of these six, three subjects scored 1, notably restless movements for less than half of the observation time; two of the six scored 2, notably restless movements for at least half of the observation time; and one subject (subject 2) scored 3, indicating constant restlessness. The latter was subject 2, who paced constantly at baseline also. Subjective akathisia scores were positive in six of the nine subjects (66.7%) who were capable of expressive language. Four of these subjects scored 1 (mild), and two scored 3 (moderate). NMS did not occur in any subjects.

Twenty-three individuals (67.6%) met the primary outcome response measure at the time of chart analysis, based on CGI-I ratings of ≤2 (showing Much Improved or Very Much Improved). Of the 14 subjects taking concomitant low dose atypical antipsychotics, 10 (71.4%) scored ≤2 on CGI-I.

Discussion

Loxapine in low doses of 5–15 mg/day shows promise for treatment of irritability and aggression in ASD, superior to atypical antipsychotics in terms of ease of use and lack of metabolic side effects. The usual loxapine dose range is 80–200 mg daily, whereas in the present study clinically prescribed doses were 5–15 mg daily, while one subject received 30 mg daily. In this study, in order to closely examine the rates of EPS, akathisia, TD, and NMS, we examined consecutive patients with ASD receiving loxapine, many of whom received it for ≥5 years. Fourteen subjects also received low doses of atypical antipsychotics; in many cases, prior treatments of atypical antipsychotic had been tapered as tolerated.

As is not uncommon for a tertiary referral center, rates of comorbid diagnoses with ASD were high. Although loxapine in add-on shows promise for irritability and aggression in ASD, with associated weight neutrality and possible brain-derived neurotrophic factor (BDNF) increase, it appears to effectively treat a broad spectrum of comorbidities in ASD. The high representation of bipolar disorder diagnoses in this sample likely relates to the complex nature of their treatment and the likelihood of referral to a tertiary care center.

At the time of chart analysis, subjects on long-term loxapine experienced mostly mild or no neuromotor side effects. Tremor, mostly mild, was the most common side effect observed, followed by akathisia. Out of 34 subjects, tremor occurred in 20.5% of subjects, and objective akathisia occurred in 17.6% of subjects. Although this rate may be higher than is generally observed with atypical antipsychotics alone, the promise of lesser or no weight gain or metabolic illness with loxapine is important. Out of the nine subjects capable of expressive language, 66.7% experienced mild or moderate subjective akathisia. However, akathisia findings overlap with baseline motor restlessness.

Only 1 of 34 subjects (2.9%), subject 26, developed TD on low dose loxapine. This is a significantly lower rate than the published rate of 5.5–32.4% for classic antipsychotics. Subject 26, who had TD, had received loxapine 5 mg daily for 2.5 years; however, he also received, from an outside provider before returning to our clinic, the SSRI paroxetine, which potently inhibits hepatic CYP2D6 metabolism of antipsychotics. Therefore, his resulting loxapine plasma levels were likely markedly elevated, because loxapine is metabolized by CYP2D6 in the liver. Subject 26 also had Prader–Willi syndrome, a genetic disorder with associated muscle abnormalities. Individuals with Prader–Willi syndrome who receive loxapine should, therefore, be closely monitored. To further assess the risk of TD in patients taking antipsychotics, in the future, genotyping of dopamine receptors may become a possibility, in order to assess genetic risks for TD (Stahl 2013).

Overall, low dose loxapine was well tolerated over the long term, with mostly mild adverse events recorded. It also likely improved behavioral outcomes: 67.6% of subjects met the primary outcome response measure at the time of chart analysis, based on CGI-I ratings of ≤2 (showing Much Improved or Very Much Improved). Of the 14 subjects taking concomitant atypical antipsychotics, 71.4% scored ≤2 on CGI-I. This is an important finding because of the potential for loxapine as an add-on to the atypical antipsychotic regimen or to spare atypical antipsychotic use. As stated previously, loxapine's unique profile of D2 versus 5-HT2 blockade may allow it to augment atypical antipsychotic treatment in partially responsive patients. Adding loxapine may be a more cost-effective choice than increasing the dose of the more costly SGA, especially if loxapine is less likely to cause both EPS and weight gain. Potentially, loxapine could, therefore, reduce health care costs while improving quality of life in ASD patients.

Weaknesses of the study include the small number of subjects, the unblinded design, sample heterogeneity, the use of concomitant medications with all subjects, and the reality that every participant had comorbid diagnoses. Nevertheless, our sample studied naturalistic outpatient cases, almost half of whom received loxapine for >5 years. Without the concomitant psychotropic medications, including atypical antipsychotics, valproic acid, paroxetine, and sertraline, rates of observed neuromotor side effects associated with loxapine would likely be lower. Individuals with bipolar disorder were included in this study, because they could be more likely than those with other comorbid psychiatric illnesses to develop TD when treated with classic antipsychotics.

Our findings suggest that low dose loxapine is safe for long-term use in ASD, and warrant double-blind, placebo-controlled trials of loxapine that test its efficacy in improving ASD symptoms, frequency of adverse events, and possible mechanisms of action.

Conclusions

The low rates of neuromotor side effects observed confirm that loxapine resembles an atypical antipsychotic in low doses in individuals with ASD, with a much lower rate of TD than occurs with full doses of classic antipsychotics. Neuromotor side effects are more likely in individuals with altered muscle biology. Concomitant prescribing of SSRIs increases the risks of serious neuromotor side effects including TD, and is discouraged. Larger, controlled studies of low dose loxapine in ASD are warranted.

Clinical Significance

Low dose loxapine, which has a unique antipsychotic profile, appears tolerable and to significantly improve treatment options for irritability and aggression in the longer term in ASD.