A group of Taiwan scientists from the Institute of Biomedical Sciences, Academia Sinica together with psychiatrists from 44 local hospitals recently successfully identified a gene that is associated with response to lithium maintenance treatment in bipolar I disorder in patients of Han descent. Their research was published in the prestigious medical journal The New England Journal of Medicine on January 9, 2014.

         The study suggests that three genetic variants in the GADL1 gene are useful as biomarkers for predicting the response to lithium maintenance treatment in patients of Asian ancestry who have bipolar I disorder. These variants are rare in persons of European and African ancestry, but it is possible that other variants in GADL1 may influence the response to lithium therapy in these populations.

         Bipolar I disorder is a mood disorder characterized by recurrent episodes of mania and a proportion of patients also experience depressive episodes with a high risk of suicide. The disease often seriously interferes with patients' personal lives and socio-occupational function, as well as their families’ quality of life.

         Currently, lithium is the first-line drug of choice for maintenance treatment of bipolar disorder and reduces risk of relapse and suicide. If bipolar patients take lithium regularly according to doctors' prescriptions (good drug adherence) and their blood levels of lithium can be kept within the therapeutic range (no less than 0.5 mmol per liter), over the long term approximately 80% have at least a partial response, and 30% show an excellent response with complete symptom remission in populations of patients of European descent. However, these statistics mean that there is still a considerable proportion of bipolar patients do not have a response to lithium preventive treatment.

         Although lithium has been used for nearly 50 years, no clinical predictors or genetic markers have been found that have sufficient sensitivity to accurately identify bipolar patients who will respond to lithium prophylaxis treatment in the clinic. Identification of such predictors (known as biomarkers) is of great importance as alternative mood stabilizing drugs may be used for patients who are predicted to have no response to lithium.

         In this study, recruitment of bipolar patients and assessment of lithium effects were conducted by psychiatric nurses and psychiatrists led by Dr. Andrew Cheng at Academia Sinica and Dr. Chau-Shoun Lee at Department of Psychiatry, Mackay Memorial Hospital, Taipei, Taiwan. Genomewide association study and genotyping and sequencing were conducted by Dr. Chien-Hsiun Chen and Drs Ming-Ta Michael Lee and Jer-Yuarn Wu under the leadership of Dr. Yuan-Tsong Chen, at the Institute of Biomedical Sciences, Academia Sinica.

         The investigators first constructed a life chart with graphic depiction of lifetime clinical course and drug treatment history (mood stabilizers and psychotropics) for a total of 1761 patients on the basis of integrated information gathered from direct interviews with the patients and their family members, interviews with psychiatrists in charge, and a thorough medical-chart review. Based on this life chart, two independent subgroups of patients were identified to have received lithium maintenance treatment with good adherence for at least two years. The investigators then assessed the response to lithium treatment among these patients using the Alda Scale developed by Dr. Martin Alda and colleagues in Canada.

         A genomewide association study was carried out among a subgroup of 294 patients. The single nucleotide polymorphisms (SNPs, variations in a gene among individuals) showing the strongest association with response to lithium were then tested for association in another subgroup of 100 patients (replication study), and further tested in a follow-up series of 24 patients who had received lithium monotherapy for at least 2 years. The life charts of these 24 patients indicated that each had had a history of good adherence to mood stabilizers other than lithium before they initiated lithium monotherapy, but had had an unsatisfactory response to such mood stabilizers.

         Two SNPs of the glutamate decarboxylase-like protein 1 (GADL1) gene showed the strongest associations in the genomewide association study (P = 5.50 x 10-37 and 2.52 x 10-37, respectively) and in the replication set of 100 patients (P = 9.19 x 10-15 for each SNP). These two SNPs had a sensitivity of 93% for predicting lithium response (i.e., for 100 responders the two SNPs accurately predicted 93) and completely differentiated between patients with a good response and those with a poor response in the follow-up cohort. A third variant, a 1-base deletion in intron 8, identified from re-sequencing of GADL1 also strongly associated with the response to lithium treatment.

         The physiological function of GADL1 protein remains unknown, but may be similar to that of glutamate decarboxylase (GAD) which is involved in the decarboxylation of glutamate and other amino acids, and is also a key enzyme in the biosynthesis of γ-aminobutyric acid (GABA). Both GADL1 and GAD are expressed in the human brain, where glutamate acts as the primary excitatory neurotransmitter and GABA is a major inhibitory neurotransmitter. Our study suggests the importance of the glutamate pathway in bipolar disorder and the possibility that lithium may affect glutamatergic neurotransmission.

         Nearly half of the 1761 patients with bipolar I disorder in this study (47.2%) carry the response allele T in one variant of GADL1, a prevalence similar to that in the general Han Chinese population, suggesting that approximately half of the patients with bipolar I disorder in Taiwan may benefit from lithium therapy. The researchers suggest that the confirmation of the findings via clinical study based on the known GADL1 allele is warranted in the near future.

         The study was financially supported by Academia Sinica and National Science Council, Taiwan.

         The full article entitled "Variant GADL1 and response to lithium therapy in bipolar I disorder" is available at the The New England Journal of Medicine journal website at: http://www.nejm.org/doi/full/10.1056/NEJMoa1212444