This randomised double-blind study was conducted to explore the efficacy of lamotrigine for the prevention of relapse and/or recurrence of manic, hypomanic, mixed or depressive episodes in Chinese patients with bipolar I disorder. The current study was based on two earlier pivotal studies [14, 15], which led to the approval of lamotrigine in the global markets including US and EU (at a target dose of 200 mg/day) [5, 7].
The number of patients who reached TIME (primary endpoint) was slightly lower in the lamotrigine group than in the placebo group; however, this difference was not significant. In addition, median survival time was slightly higher for the lamotrigine group (188 days) than that of the placebo group (183 days) but superiority of lamotrigine was not demonstrated. Similarly, all secondary efficacy endpoints except for body weight (TIMan, TIDep, TIME-SIS, CGI-S, CGI-I, HAMD, YMRS and GAS) did not show significant differences between the lamotrigine and placebo groups. Body weight showed a significantly greater decrease among the lamotrigine group. These findings contrast with those from the previous pivotal studies in international populations. In both of these studies, lamotrigine was superior to placebo in prolonging TIME and TIDep but not TIMan [14, 15]. The differences in the changes from baseline in HAMD between the lamotrigine and placebo groups in pivotal studies were also significant and the median survival time was significantly longer with lamotrigine versus placebo (85 vs 58 days [p=0.03] and 200 vs 93 days [p=0.003]) [14, 15]. The possible reasons for the differences between these results from the pivotal studies and the current findings merit consideration.
The lack of significant efficacy in the current study population could be because the patients included had milder disease compared with the previous study populations despite mostly similar inclusion and exclusion criteria [17, 16, 14]. There is evidence in the literature that disease characteristics at baseline have a strong impact on the treatment outcome of patients with bipolar I disorder. Patients who have more prior mood episodes are known to have a slower response to treatment and a higher risk of relapse/recurrence of further episodes than patients who have fewer prior mood episodes [25–27]. Patients who present with severe mood symptoms at baseline generally respond slowly to treatment and are likely to present with residual mood symptoms during recovery after the initial treatment; the presence of residual symptoms is a major risk factor for relapses/recurrences during maintenance treatment [28–31, 25, 32]. The patients in this Chinese study had a shorter bipolar I disorder disease history and milder mood symptoms at baseline compared with previous pivotal studies. In the present study, the mean CGI-S score at screening was 3.6, which is lower than reported in the pivotal trials (4.3 and 4.4) [14, 15].
Furthermore, patients in the present study also had a lower number of depression and mania/mixed episodes in the preceding 3 years and a lower rate of suicide attempts compared with the pivotal studies [14, 15]. Therefore, the current study population could be expected to have a lower risk of relapse/recurrence of mood episodes during the RD phase than the populations in the pivotal studies based on the severity of disease at baseline.
Additional factors may also have contributed to the lack of a significant effect of lamotrigine in the main study population. Patients in both the lamotrigine and placebo groups experienced a low number of mood episodes during the 36-week RD phase, as expected as a consequence of their relatively mild disorder. The dropout rate was also higher than anticipated (56% compared with the assumed 20%) further reducing the number of patients available to experience an event. Consequently, superiority of lamotrigine over placebo in preventing the relapse/recurrence of mood episodes could not be demonstrated. The duration of the RD phase within the present study was also comparatively shorter (36 weeks) compared with the previous studies (approximately 78 weeks) [14, 15]; a longer duration of study may have resulted in a higher number of mood episodes allowing estimation of time to an event and potential demonstration of a significant effect. In a previous phase I study in healthy Chinese volunteers, lamotrigine demonstrated a comparable pharmacokinetic profile to that in the international population [5, 33]. Therefore, differences in genetics and ethnicity are not thought to play a major role in the differences in results between this study in Chinese patients and the studies conducted in international populations.
The post hoc analyses aimed to explore the potential impact of disease severity at baseline on the lack of a significant effect of lamotrigine by evaluating the response i.e. the primary endpoint (TIME) in a subgroup of patients with moderate/severe disease at baseline using the criteria of CGI-S or HAMD/YMRS thresholds. These post hoc data showed a statistically significant difference in favour of lamotrigine versus placebo in patients with moderate/severe mood symptoms (baseline CGI-S score ≥4 or HAMD ≥18/YMRS ≥10). Although the data from the post hoc analyses need to be interpreted with caution, these results are more consistent with the observations from previously mentioned international studies [14, 15].
Overall, lamotrigine 200 mg/day showed a comparable safety and tolerability profile to placebo and led to a lower number of AEs compared to the prior studies outside China [14, 15, 34, 17]. The number of patients who experienced serious TEAEs and AEs leading to withdrawal from the study were similar in both placebo and lamotrigine groups. No new or unexpected safety signals emerged based on the AEs, laboratory analyses, and monitoring of vital signs and ECGs.
This study has a number of limitations. The study may have benefitted from longer lead-in requiring patients to be stable for longer than 4 weeks. Diagnosis of bipolar disorder in this study was made without using standardised measures (e.g. Structured Clinical Interview for DSM Disorders). Moreover, there was no active comparator. The use of concomitant medications during the treatment phase, including lamotrigine and valproate sodium in the RD phase, may have impacted the results.