gms | German Medical Science

Introduction: The multicenter randomized controlled phase-III trials ONLINE-TICS [1] and CANNA-TICS [2] investigate two different interventions for the treatment of Tourette syndrome (TS) and chronic tic disorders. ONLINE-TICS investigated the use of internet-delivered Comprehensive Behavioral Intervention for Tics (iCBIT) (n = 67) compared to face-to-face CBIT (n = 24) and a placebo (sham intervention), defined as online psychoeducation (n = 70). CANNA-TICS was a phase-III drug trial that aimed to demonstrate efficacy of the cannabis extract nabiximols (n = 64) versus placebo (n = 33). A re-evaluation of both trials may generate useful evidence since the trial designs and populations appeared to be similar, and the data documentation in both trials was identical. The re-analysis of the individual patient data from the two trials may show more detailed insight into different treatments for tics and supports the discussion about trial designs for future tic studies.

Methods: The trial designs were compared substantively following the PICO(S) scheme [3]. Descriptive analyses of baseline variables and the main outcomes were performed to explore similarities and differences between trials using a two-sided t-test for continuous data and a Chi-squared test for dichotomous data. Since the Yale Global Tic Severity Scale-Total Tic Score (YGTSS-TTS) was used as the primary endpoint measure in both trials, the YGTSS-TTS was re-analyzed exploratory over the trial courses.

Results: The comparison of the design aspects by the PICO(S)-scheme revealed differences between the trials, for example, different interventions were tested and the primary analysis of the YGTSS-TTS was defined differently. The YGTSS-TTS was analyzed continuously in the ONLINE-TICS trial by estimating least square mean (LSM) differences using a linear mixed model approach. In contrast, in the CANNA-TICS trial, a dichotomous responder criterion [4] was defined and analyzed using a Mantel-Haenszel test for the risk difference (RD) between treatments. Regarding common variables, the trial populations differed at baseline, particularly for tic severity according to the YGTSS-TTS (ONLINE-TICS: 24.4 (8.1), CANNA-TICS: 28.8 (8.6), p <.0001). Even though the YGTSS-TTS at baseline differs between the trials, the change of the YGTSS-TTS over the course of the trials is similar. Re-analysis of the trial data showed similar point estimates for both estimators, RD and LSM difference, and comparable variability.

Discussion and conclusion: Although both trials showed promising results for the efficacy of the respective interventions in treating TS and chronic tic disorders over the trial courses, both failed their primary objective. In both trials observed results deviated slightly compared to the expectations in the sample size calculation. Since the primary analysis was defined differently for each trial, treatment effect estimates may vary depending on the definition of the primary endpoint measure and the statistical analysis method. This allows the discussion about whether and under which circumstances the primary endpoint should be defined as a dichotomous or continuous measure.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.