A New Class of Antipsychotic Drugs

December 1, 2008

The development of new, more effective antipsychotics with fewer adverse effects (eg, extrapyramidal symptoms, tardive dyskin­esia, metabolic syndrome) is paramount.

The development of new, more effective antipsychotics with fewer adverse effects (eg, extrapyramidal symptoms, tardive dyskin­esia, metabolic syndrome) is paramount. In most patients with schizophrenia, the disease is resistant or only partially responsive to treatment with available antipsychotics. In addition, the adverse effects of both typical and atypical antipsychotics add to the difficulty of obtaining a satisfactory treatment response.

Given that the neurobiology of schizophrenia is poorly understood, the development of new treatments is primarily based on current pharmacological models of schizophrenia, such as dopaminergic and serotoninergic models. In addition to these models of neurotransmission, glutamatergic neurotransmission has been implicated in the pathophysiology of schizophrenia.1,2 Because the N-methyl d-aspartic acid (NMDA) receptor, a subtype of the ionotropic glutamate receptor, plays an important role in neurodevelopment and cognition, deficits in NMDA neurotransmission can potentially account for developmental risk factors and cognitive impairments in schizophrenia.


The involvement of the NMDA sys­tem in schizophrenia is further evidenced by the effects of the psy­­­­­­­choto­­­­­­­mimetic drug phencyclidine and the dissociative anesthetic, ketamine, both of which are NMDA-receptor antagonists. The psychotic symptoms induced by these NMDA antagonists involve not only positive symptoms, as seen with dopaminergic agonists, but negative symptoms and cognitive deficits as well.3 These findings support the possibility that symptoms observed in patients with schizophrenia could arise through attenuated NMDA receptor-mediated neurotransmission. It can be further postulated that the psychotomimetic effects may not be limited to noncompetitive antagonists but could result from any dysfunctional attenuation of the NMDA receptor-mediated neurotransmission.

NMDA involvement in schizophrenia
Two lines of study support the pharmacological model of NMDA hypofunction as being involved in schizophrenia. The first line of evidence comes from findings of clinical studies that indicate that new NMDA-enhancing agents are beneficial in the treatment of schizophrenia symptoms. These agents serve as coagonists at the NMDA receptor, which bind to the glycine (and/or d-serine) binding site concurrently with separate occupancy of the glutamate binding site (Figure). Occupancy of both glutamate and “glycine” sites is essential for the activation of the NMDA receptor. Amino acids acting on either site are coagonists; just as d-serine or glycine cannot activate the NMDA receptor, neither aspartate nor glutamate alone can activate the NMDA receptor without the co-occupancy of the glycine “modulatory” site.

Considering that agonists of the NMDA recognition site are excitotoxic, molecules that act on the obligatory NMDA-glycine site are promising therapeutic candidates. Several agents that directly or indirectly enhance the NMDA function via the activation of the NMDA-glycine site have been tested to determine their effectiveness in the treatment of schizophrenia. The Table summarizes the mechanisms of action of these agents. These agonists of the NMDA-glycine site include the simple amino acids glycine, d-serine, d-alanine, and the partial agonist d-cycloserine.4-7 Most of the double-blind placebo-controlled studies of the NMDA-enhancing agents showed significant improvement in multiple symptom domains of schizophrenia, including positive, negative, cognitive, and depressive symptoms. In addition, the effect of d-cycloserine is limited to negative symptoms and the effect is at times negative.6

Because glycine transporter 1 (GlyT1) plays a pivotal role in maintaining the concentration of glycine within NMDA synapses at a subsaturating level, and the anatomical distribution of GlyT1 is parallel to that of the NMDA receptor, another approach to enhance NMDA neurotransmission is by blocking the reuptake of glycine through GlyT1. This is analogous to the rationale behind using a serotonin reuptake inhibitor to enhance serotonergic neurotransmission. In support of the critical role that GlyT1 plays in NMDA neurotransmission, the GlyT1 inhibitor, a sarcosine (N-meth­yl­­gly­cine) analogue, N[3-(49-fluoro­phenyl)3-(49 phenylphenoxy)propyl] sarcosine, and the GlyT1 knock­down mutation have been shown to enhance NMDA neurotransmission.8 Sarcosine, which is an en­dogenous inhibitor of GlyT1, has shown clinical efficacy when added to typical and atypical antipsychotics, and it has also shown efficacy when administered alone, thereby supporting its NMDA-enhancing and anti­psychotic function.9

The second line of inquiry comes from a series of genetic linkage and association studies that support hypofunction of NMDA neurotransmission in schizophrenia, a common defect at glutamatergic and other synapses, including the NMDA synapse.10 NMDA neurotransmission is influenced to varying degrees by most, if not all, of the putative susceptibility genes including the proline dehydrogenase gene, the dysbindin gene, the neuregulin gene 1, the disrupted-in-schizophrenia gene 1, the V-akt murine thymoma viral oncogene homolog gene 1, the regulator of G-protein signaling gene 4, and the metabotropic glutamate receptor gene 3. These risk genes include the enzyme that catabolizes d-serine and d-alanine, d-amino acid oxidase (DAAO), and its primate-specific activator G72; they are directly involved in NMDA neurotransmission (Figure). DAAO appears to be the critical determinant of d-serine levels, since its activity correlates inversely with d-serine levels both regionally and developmentally. Facilitation of G72 function up-regulates DAAO activity, enhances metabolism of d-serine and d-alanine, and attenuates NMDA neurotransmission.

Over the past 5 years, more than 10 studies have demonstrated the association of G72 with schizophrenia or bipolar disorder.11 This is an intriguing association, given recent findings that serum and cerebrospinal fluid levels of d-serine are reduced in patients with schizophrenia.12

NMDA-enhancing treatments
In contrast to early studies, which suggested that the therapeutic effects of NMDA-enhancing agents added to antipsychotics were mainly on negative symptoms, more recent findings suggest that the effects extend to positive, cognitive, and depressive symptoms when compared with placebo alone.4-7,9 Overall, the effect sizes are in the small to medium range and are similar to those seen with atypical antipsychotics.13

The wide spectrum of the therapeutic effects of the NMDA-enhancing agents would imply that their efficacy may not be limited to treatment of schizophrenia but that they can potentially be useful for other mental disorders involving psychotic, cognitive, or mood manifestations. In fact, efficacy results for these agents have led to recent studies to evaluate them as sole pharmacotherapeutic agents, as exemplified by a recent trial of sarcosine in antipsychotic-free patients.14

Effects on negative and cognitive symptom domains
Negative symptoms are often refractory to antipsychotic treatment with the exception of clozapine (and, to a lesser degree, the other atypical antipsychotics). The NMDA-enhancing agents have a therapeutic effect on the negative symptoms in patients with chronic schizophrenia who have been receiving stable doses of antipsychotics before the initiation of the NMDA-enhancing agents.4-9,14 The improvement in negative symptoms by NMDA-enhancing agents can have critical implications in the long-term functional outcome of patients with schizophrenia since negative symptoms are one of the main causes of disability and poor outcome.13

Neurocognitive impairment is similarly associated with the limited outcome and poor functioning in patients with schizophrenia.15 It has been suggested that cognitive impairment may have a stronger relationship with poor outcome than other symptom domains.13 While there is a significant effect of NMDA-enhancing agents on scores on the Positive and Negative Syndrome Scale (PANSS) cognitive-subscale, most studies have been short-term trials that looked at symptom reduction and thus lack an assess­ment of cognitive domains by neurocognitive testing.5,6,9 Taken together, it is premature to draw any conclusions regarding the cognition-improving effect of the NMDA-enhancing agents. It is also critically important to conduct long-term trials and neu­­ro­cog­ni­tive studies to determine wheth­er the symptom reduction, particularly the reductions in negative and cognitive symptoms, can improve and sustain the cognition and long-term functional outcome of patients who have schizophrenia.

Effect of different NMDA-enhancing agents
Given that the NMDA-enhancing agents are not a homogeneous group, their varied efficacy is not surprising. d-Serine and glycine treatment has a more comprehensive symptom-improvement profile than does d-cycloserine. Specifically, the effects of d-serine and glycine are greater than those of d-cycloserine on negative symptoms.4-7,9,14 This is probably because d-serine and glycine are full agonists whereas d-cycloserine is a partial agonist that cannot fully activate the NMDA receptor. When combined with a substantial NMDA agonist, d-cycloserine behaves like an antagonist, thereby worsening the psychosis. In fact, the results of d-cycloserine trials are mixed. Consistent with its role as a full agonist and its low CNS bio-availability by peripheral administration, higher doses of glycine are more effective for negative symptoms than the lower doses used in earlier studies.4

d-Serine significantly improves negative, positive, cognitive, depressive, and total psychopathology. While the effects of d-serine are similar to those of glycine in negative, cognitive, depressive, and total psychopathology, d-serine, but not glycine, also improves positive symptoms. This can be explained by the better central bioavailability of serine. Future studies on optimal glycine and d-serine dosing can clarify whether the difference is caused by central bio-availability. d-Alanine, another full agonist of the NMDA-glycine site, has thus far been tested only by our group and has shown a comprehensive efficacy profile similar to that of d-serine.6

Although there are several GlyT1 inhibitors in preclinical development, only sarcosine has been clinically tested in 5 studies: 3 studies in stable patients with chronic schizophrenia, 1 in patients with acute exacerbation, and 1 in which sarcosine was used as the sole antipsychotic agent.9,14,16-19 In trials with stable chronically ill patients who received antipsychotics other than clozapine, the efficacy of sarcosine was similar to that of d-serine and d-alanine.9,19 In addition, 2 recent studies suggest that sarcosine, more than d-serine, can be of benefit to both acutely ill and chronically stable patients with schizophrenia.18,19 Compared with those in the placebo group, patients in the sarcosine group rather than patients in the d-serine group were more likely to respond. Both sarcosine and d-serine are superior to placebo in score-changing rates of the PANSS-total and Quality-of-Life scales. In addition, sarcosine distinctly elicited better response rates than control when administered as the sole antipsychotic agent, particularly in antipsychotic-naive patients.14 Taken together, these findings indicate the interesting possibility that the GlyT1 inhibitor may be more effective than NMDA-glycine site agonists at the doses being tested for the treatment of schizophrenia. One possibility is that the anatomical distribution of GlyT1 is more relevant than the NMDA receptor per se to the circuitries involved in schizophrenia. Nevertheless, further confirmation and parallel comparison are required to determine effective dosing ranges and to compare the effectiveness of the full agonist of the NMDA-glycine site with the GlyT1 for treating schizophrenia. Caution is needed in interpreting these exploratory studies; all but 1 of which were performed by our group with similar design and inclusion/exclusion criteria.

d-Cycloserine has a very narrow therapeutic window. Goff and colleagues20 found that doses higher than 100 mg worsened the symptoms of schizophrenia, as well as the symptoms of patients who received clozapine. This raises an intriguing possibility that part of clozapine’s action may involve the NMDA-glycine site. This hypothesis is consistent with the negative findings for glycine, d-serine, and sarcosine in patients who were treated with clozapine.17

Almost all the studies of the NMDA-enhancing agents were performed with patients on stable doses of antipsychotics. NMDA-enhancing agents as a whole are effective for patients who receive antipsychotics other than clozapine, including both typical and atypical antipsychotics, but not in patients treated with clozapine. Scores on the Clinical Global Impression Improvement Scale were between 2 and 3, indicating a mild to moderate improvement that could be clinically observed. The addition of NMDA-enhancing agents to atypical antipsychotics (risperidone and olanzapine) resulted in significant improvements in total psychopathology, negative symptoms, cognitive symptoms, and depressive symptoms.9,18,19 These findings underscore the significance of NMDA enhancement as a new therapeutic approach in an era in which atypical antipsychotics are becoming the primary drugs for most patients who have schizophrenia. We theorize that this may involve a synergistic therapeutic effect of the atypical antipsychotics and NMDA-enhancing agents, providing an intriguing augmentation strategy for patients whose disorder is resistant or only partially responsive to atypical antipsychotics.

Adverse effects and safety
Except for GI upset and nausea in some glycine trials, other adverse effects have not been significantly different between treatment with placebo and with the NMDA-enhancing agents.17-21 Adding NMDA-enhancing agents to other antipsychotics similarly showed a good safety profile and did not worsen the adverse effects of concomitant antipsychotics, which are mediated by D2, 5-HT2, histamine, muscarinic, or other receptors. Nevertheless, long-term studies are needed to thoroughly evaluate the safety of the NMDA-enhancing agents. If confirmed by results from additional studies, these would be safe antipsychotic agents devoid of the adverse effects of extrapyramidal symptoms, tardive dyskinesia, and metabolic syndrome. These advantages will help a substantial portion of patients who experience serious adverse effects while taking the existing antipsychotics.

In fact, given that NMDA-enhancing agents were found to significantly improve extrapyramidal symptoms on both the Simpson-Angus Scale and Abnormal Involuntary Movement Scale measurements, it is possible that NMDA-enhancing agents may produce both better and safer therapeutic results. A recent study found that coadministration of an NMDA-enhancing agent can reduce the dosage of an antipsychotic required to reach the same level of symptom reduction in schizophrenia while minimizing adverse effects.21 Therefore, the add-on strategy may actually lead to a reduction of dose-related adverse effects associated with currently used antipsychotics. Alternatively, both D2 blockers and NMDA enhancers may address different dimensions of the illness, with their effects being additive, or even synergistic, without negating each other.

While the results of the trials with NMDA-enhancing agents are encouraging, most studies involve small series. The optimal dosage for each individual agent is unknown, except for d-cycloserine, which has marginal therapeutic effects in a narrow therapeutic window. Thus far, NMDA-enhancing agents have been tested mainly as add-on treatment for patients on stable antipsychotic regimens. Our belief that the NMDA-enhancing agents should be tested as monotherapy for psychosis is encouraged by initial findings, particularly in antipsychotic-naive patients.14 In addition to efficacy in treating schizophrenia, the addition of NMDA-enhancing agents has been shown to be beneficial in Alzheimer disease, posttraumatic stress disorder, autism, and phobias.22-25 In theory, any neuro-psychiatric disorder involving attenuated NMDA function could benefit from the strategy of enhancing the NMDA function.

NMDA hypofunction may be more prominent early in the disease process and, consequently, the disease may be more amendable by early use of NMDA-enhancing agents. Clearly, the new therapeutic approach of enhancing NMDA neurotransmission is still in its early development. Nonetheless, NMDA-enhancing agents improve a wide spectrum of symptoms of schizophrenia with minimal adverse effects and a satisfactory safety profile. Among the 3 agents analyzed, the full agonists, d-serine and glycine, and the GlyT1 inhibitor, sarcosine, have a better symptom profile than the partial agonist, d-cycloserine; there are preliminary suggestions that the GlyT1 inhibitor is more effective than the glycine site agonist.

Given that hypo-NMDA neurotransmission plays a role in neurotoxicity and the developmental risks of schizophrenia, correction of NMDA hypofunction early on could not only benefit the symptoms but also improve the disease course. This notion is supported by our recent finding that antipsychotic-naive patients benefit more from sarcosine therapy than patients who have been previously treated with an antipsychotic.

In the future, it would be interesting to explore the role of NMDA-enhancement treatment, not only as symptom treatment but also as illness prevention and long-term function outcome. Moreover, larger placebo-controlled studies are warranted to compare NMDA-enhancing agents with antipsychotics to determine their respective strengths and weaknesses. The addition of NMDA-enhancing agents to the antipsychotic armamentarium should enlarge our range of choices for individually tailoring or combining agents for additive effects or synergism. Innovative application of other potential targets to enhance NMDA neurotransmission may be beneficial not only for patients with schizophrenia but for those with other neuropsychiatric conditions (Figure).


Drugs Referenced in this article

Clozapine (Clozaril)
d-Cycloserine (Seromycin)
Ketamine (Ketanest, others)
Olanzapine (Zyprexa)
Risperidone (Risperdal)

1. Tsai G, Coyle JT. Glutamatergic mechanisms in schizophrenia. Annu Rev Pharmacol Toxicol. 2001;42: 165-179.
2. Lindsley CW, Shipe WD, Wolkenberg SE, et al. Progress towards validating the NMDA receptor hypofunction hypothesis of schizophrenia. Curr Top Med Chem. 2006;6:771-785.
3. Krystal JH, Karper LP, Seibyl JP, et al. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans: psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch Gen Psychiatry. 1994;51:199-214
4. Javitt DC, Silipo G, Cienfuegos A, et al. Adjunctive high-dose glycine in the treatment of schizophrenia. Int J Neuropsychopharmacol. 2001;4:385-391.
5. Tsai G, Yang P, Chung LC, et al. D-Serine added to antipsychotics for the treatment of schizophrenia. Biol Psychiatry. 1998;44:1081-1089.
6. Tsai GE, Yang P, Chang YC, Chong MY. D-Alanine added to antipsychotics for the treatment of schizophrenia. Biol Psychiatry. 2006;59:230-234.
7. Goff DC, Tsai G, Levitt J, et al. A placebo-controlled trial of D-cycloserine added to conventional neuroleptics in patients with schizophrenia. Arch Gen Psychiatry. 1999;56:21-27.
8. Tsai G, Ralph-Williams RJ, Martina M, et al. Gene knockout of glycine transporter 1: characterization of the behavioral phenotype. Proc Natl Acad Sci U S A. 2004;101:8485-8890.
9. Tsai G, Lane HY, Yang P, et al. Glycine transporter I inhibitor, N-methylglycine (sarcosine) added to antipsychotics for the treatment of schizophrenia. Biol Psychiatry. 2004;55:452-456.
10. Harrison PJ,Weinberger DR. Schizophrenia genes, gene expression, and neuropathology: on the matter of their convergence. Mol Psychiatry. 2005;10:40-68.
11. Chumakov I, Blumenfeld M, Guerassimenko O, et al. Genetic and physiological data implicating the new human gene G72 and the gene for D-amino acid oxidase in schizophrenia. Proc Natl Acad Sci U S A. 2002; 99:13675-13680.
12. Hashimoto K, Engberg G, Shimizu E, et al. Reduced D-serine to total serine ratio in the cerebrospinal fluid of drug naive schizophrenic patients. Prog Neuropsychopharmacol Bio Psychiatry. 2005;29:767- 769.
13. Altamura AC, Bobo WV, Meltzer HY. Factors affecting outcome in schizophrenia and their relevance for psychopharmacological treatment. Int Clin Psychopharmacol. 2007;22:249-267.
14. Lane HY, Liu YC, Huang CL, et al. Sarcosine (N-methylglycine) treatment for acute schizophrenia: a randomized, double-blind study. Biol Psychiatry. 2008;63:9-12.
15. Green MF.What are the functional consequences of neurocognitive deficits in schizophrenia? Am J Psychiatry. 1996;153:321-330.
16. Lechner SM. Glutamate-based therapeutic approaches: inhibitors of glycine transport. Curr Opin Pharmacol. 2006;6:75-81.
17. Lane HY, Huang CL, Wu PL, et al. Glycine transporter I inhibitor, N-methylglycine (sarcosine), added to clozapine for the treatment of schizophrenia. Biol Psychiatry. 2006;60:645-649.
18. Lane HY, Liu YC, Huang CL, et al. Double blind comparison of sarcosine (N-methylglycine), D-serine and placebo treatments for chronic schizophrenia. Presented at: Annual Meeting of American College of Neuropsychopharmacology; December 9-13, 2007; Boca Raton, FL.
19. Lane HY, Chang YC, Liu YC, et al. Sarcosine or D-serine add-on treatment for acute exacerbation of schizophrenia: a randomized, double-blind, placebocontrolled study. Arch Gen Psychiatry. 2005;62:1196- 1204.
20. Goff DC,Tsai G, Manoach DS, Coyle JT. Dose-finding trial of D-cycloserine added to neuroleptics for negative symptoms in schizophrenia. Am J Psychiatry. 1995;152:1213-1215.
21. Heresco-Levy U, Javitt DC, Ebstein R, et al. D-Serine efficacy as add-on pharmacotherapy to risperidone and olanzapine for treatment-refractory schizophrenia. Biol Psychiatry. 2005;57:577-585.
22. Tsai G, Falk W, Gunther G, Coyle JT. D-Cycloserine improves cognition of Alzheimer’s disease. Am J Psychiatry. 1999;156:467-469.
23. Heresco-Levy U, Kremer I, Javitt DC, et al. Pilotcontrolled trial of D-cycloserine for the treatment of post-traumatic stress disorder. Int J Neuropsychopharmacol. 2002;5:301-307.
24. Posey DJ, Kem DL, Swiezy NB, et al. A pilot study of D-cycloserine in subjects with autistic disorder.Am J Psychiatry. 2004;161:2115-2117.
25. Ressler KJ, Rothbaum BO, Tannenbaum L, et al. Cognitive enhancers as adjuncts to psychotherapy: use of D-cycloserine in phobic individuals to facilitate extinction of fear. Arch Gen Psychiatry. 2004;61:1136- 1144.