The aim of our study was to compare the difference in serum levels of anti-glutamate receptor antibodies between TRS patients and nTRS patients, and to discuss the effect of clozapine on antibody levels. There are few studies on anti-glutamate receptor antibody levels in TRS. Only one study directly measured the serum anti-NMDAR antibody concentration in TRS patients using ELISA and the results of our study were consistent with that, namely, increased anti-NMDAR antibody levels were associated with antipsychotic drug resistance[35]. The concentration of several other serum anti-glutamate receptor antibodies was also measured for the first time in this study, and the results were similar to those of anti-NMDAR antibodies, that is, TRS patients had higher serum anti-glutamate receptor antibody levels compared with nTRS patients. In addition, this study showed that compared patients of the TRS-nC groups, patients of the TRS-C group had lower levels of anti-glutamate receptors, milder positive symptoms, and partially better cognitive function. Taken together, these findings suggest that TRS may be associated with higher anti-glutamate receptor antibody levels, and clozapine can alleviate clinical symptoms and cognitive deficits that are difficult to treat with other antipsychotics by reducing the level of anti-glutamate receptor antibodies.
Numerous studies have shown that NMDAR dysfunction may be involved in the pathogenesis of schizophrenia and is widely associated with psychotic symptoms and cognitive impairment[44]. In addition to NMDAR dysfunction, AMPAR dysfunction mentioned in this study has been associated with selective attention and memory impairment in schizophrenia[45]. The activation of mGluR3 in the prefrontal cortex plays an important role in working memory and cognition[46], and variation in GRM3, the gene encoding mGlu3, has been shown to be associated with schizophrenia and cognitive deficits[47]. mGluR5 knockout mice show defects in prepulse inhibition (a control deficit model of schizophrenia)[48]. mGluRs agonists have now become new targets for antipsychotic drug development. In preclinical studies, both mGluR3 and mGluR5 agonists have shown some degree of therapeutic efficacy on schizophrenia, including the improvement of negative symptoms and cognitive deficits that are difficult to be improved by current antipsychotics[49, 50]. In line with our hypotheses, TRS patients have higher levels of various types of anti-glutamate receptor antibodies that cause dysfunction of glutamate receptors, so symptoms and cognitive deficits are more severe than in nTRS patients, while current antipsychotic drugs mostly act on dopamine receptors and have limited effect on negative symptoms and cognitive impairment, which is an important cause of refractoriness.
The results of our study further indicate that clozapine may modulate the function of anti-glutamate receptors by reducing antibody levels, and this may be one of the therapeutic mechanisms by which clozapine differs from other antipsychotics. This finding is also supported a previous study[42]. Compared with the clozapine-naive group, the clozapine-treated group had significantly lower immunoglobulin levels[42]. And this cross-sectional study also reported a significant association between the duration of clozapine treatment and the degree of reduction in serum IgG levels[42]. Indeed, in recent years, some studies have linked clozapine to SAD, leading researchers to speculate that this may be one of the therapeutic mechanisms of clozapine[42, 43, 51]. Unfortunately, there is a lack of relevant studies in large samples, with only isolated case reports of reduced autoantibody titers and symptom improvement in TRS patients with serum anti-NMDAR antibodies after clozapine administration[52]. The mechanism by which clozapine reduces anti-glutamate receptor antibody levels is currently unknown. It has been reported that dopamine D1 receptors are the main receptor type expressed by tonsillar B cells[53], and researchers have speculated that the underlying mechanism is related to the antagonistic effect of antipsychotic drugs on dopaminergic signaling in the germinal center[43]. However, it also means that the antagonistic effect of other antipsychotics should be stronger than that of clozapine, but the clozapine-treated group had lower antibody levels in both previous studies and in the present study, so how exactly clozapine affects antibody levels and the effect of other antipsychotics on antibodies needs to be further explored. In this study, no correlation was observed between clozapine dose and any of the anti-glutamate receptor antibody levels, probably because most patients of the TRS-C group were not treated with clozapine alone and therefore could not control for the effects of other antipsychotics; moreover, different individuals taking the same dose of clozapine, due to various factors such as metabolism, smoking, gender, and combination with other drugs, may also have different blood levels, and thus vary in the degree of alteration of antibodies[54, 55].
After Bonferroni correction, no correlation between antibody concentrations and symptom severity and cognitive function was found in any of the three groups of patients. First, as an exploratory study with a small sample size, the multiple comparison correction seems too rigorous here. And similarly, another study also measured serum NMDAR antibodies in TRS, nTRS, and treatment initiation schizophrenia (1. first-episode schizophrenia; 2. total duration of illness ≤ 3 years; and 3. within 2 weeks of initiation of antipsychotic medication), and the correlation between antibody levels and symptom severity was observed only in treatment initiation schizophrenia[35]. Long-term antipsychotic use also affects neurotransmitter mechanisms other than glutamate, making the correlation between antibody levels and symptoms no longer significant[35]. Furthermore, although declined glutamatergic signaling is thought to be one of the causes of various cognitive impairments (e.g., probabilistic reasoning, working memory) in schizophrenia[56, 57]. Nevertheless, there are a generalized deficits in cognitive function in schizophrenia that unable to establish a clear pattern[58], with numerous possible mechanisms involving many other neurotransmitters (e.g., GABA abnormalities may be associated with alterations of gamma oscillatory activity in the prefrontal cortex and caused working memory disturbances[59]). Meanwhile, in the TRS-C group, no correlation was found between either their cognitive function or symptom severity and the levels of the four antibodies, probably because of the complex mechanism of action of clozapine, which could not only affect the glutamate antibody levels, but also enhance the function of glutamate receptors such as NMDAR through different mechanisms, including blocking the uptake of glycine, increasing the occupancy of glycine sites, and interaction with mGluR5[60–63]. Moreover, it can also act on other receptors, including muscarinic, histamine, and serotonin receptors.
The differences in clinical symptoms and cognitive function among the three groups were also compared and analyzed on this study. The results indicated that, compared with patients with nTRS, patients with TRS have more severe symptoms. Moreover, clozapine can mainly improve positive symptoms and partially alleviate cognitive impairment, and have a limited effect in the treatment of negative symptoms and cognitive impairment in some domains (e.g., attention and vigilance, visual learning). There have been longitudinal studies comparing the efficacy of clozapine and chlorpromazine in patients with schizophrenia, and clozapine also showed some degree of improvement in negative symptoms over chlorpromazine[64, 65]. However, these studies also found comparable efficacy with clozapine versus chlorpromazine at the endpoints of the studies. The advantage of clozapine is its better tolerability[65]. Consistent with our findings, a Meta-analysis comparing the efficacy of clozapine with first- and second-generation antipsychotics for TRS found that clozapine was superior for positive symptoms in both the short and long term, and superior for total and negative symptoms only in the short term[66]. The average duration of TRS patients in both groups included in this study was more than 10 years, and a considerable part of them were hospitalized for a long time. Therefore, clozapine did not improve their negative symptoms and general symptoms significantly, while it was still better than other antipsychotics for the treatment of positive symptoms. However, it is still unclear why the long-term benefit of clozapine is limited to the treatment of positive symptoms[66]. In this study, even though all serum anti-glutamate receptor antibodies were lower in TRS patients treated with clozapine than in TRS patients not treated with clozapine, there was no significant difference in the severity of symptoms other than positive symptoms, suggesting that in the context of a long disease course, negative symptoms are persistent and difficult to treat, and their pathophysiological mechanisms are more complex and not limited to elevated serum anti-glutamate receptor levels and decreased glutamate receptor function. Indeed, negative symptoms and cognitive impairment in schizophrenic patients are usually present before the first episode[67]. They may originate from abnormalities in early neurodevelopmental processes[68]. Although modulation of glutamate receptor function may be beneficial for negative and cognitive symptoms, replicated evidence is lacking[69]. Future therapeutic targets for negative symptoms and cognitive impairment may extend beyond the glutamate system. Several new targets have been proposed, such as microglia, the complement system and the GABA system[70].
There are some limitations of this study. Limited by the specificity of the study subjects, the sample size of the study was relatively small, and it was also difficult to recruit the three groups of subjects with matched illness duration, education, and medication. Therefore, in the data analysis, correlation analysis was performed or covariates were included to eliminate their effects. Moreover, because clozapine initiation is often delayed in clinical practice, the illness duration of subjects was long and, in consequence, there were many confounding factors that were difficult to exclude. Hence, the inclusion of healthy controls and first-episode schizophrenia patients and longitudinal study should be considered in future studies.
In summary, we found that increased anti-glutamate receptor antibody levels were associated with TRS and clozapine could reduce the antibody levels in TRS. Our findings may provide evidence to support the glutamatergic dysfunction hypothesis and the neuroimmunological hypothesis. Glutamate receptor agonists and immunotherapy can be new targets for the treatment of TRS in the future. And it is also suggested that the effect of clozapine targeting anti-glutamate receptor antibody levels may be one of the possible mechanisms underlying its superior efficacy. The use of animal models in further researches are required to understand the ability and pathway of clozapine to reduce antibody levels[41, 43]. In addition, we observed that negative symptoms and cognitive impairments seemed to be clozapine resistant, which indicate that the pathophysiology can be more complex, and besides glutamate, there are various neurotransmitters involved.