• DOI:10.1111/epi.16607
• Corpus ID: 220520465

Cannabidiol efficacy and clobazam coadministration: Where do we stand now?

@article{Lattanzi2020CannabidiolEA,
  title={Cannabidiol efficacy and clobazam coadministration: Where do we stand now?},
  author={Simona Lattanzi and Francesco Brigo and Eugen Trinka},
  journal={Epilepsia},
  year={2020},
  volume={61}
}

• S. Lattanzi, Francesco Brigo, E. Trinka
• Published 14 July 2020
• Medicine
• Epilepsia

To the Editor In this issue of Epilepsia, two articles deal with the debated relationship between cannabidiol (CBD) efficacy and concomitant clobazam (CLB) use.1,2 The differences and similarities of these studies deserve some comments to adequately appreciate their strengths and limits. The first review1 is a critical appraisal of publicly available analyses of seizure outcomes on CBD treatment as presented by the applicant and provided by the European Medicines Agency (EMA) Assessment Report.3 Data from the four pivotal randomized-controlled trials (RCTs) were first analyzed independently using negative binomial regression for the primary efficacy outcome of change in baseline seizure frequency, and then combined using a stratified fixed-effects meta-analysis. Results were presented as treatment ratios (placebo/CBD) rather than absolute changes or relative reductions in seizure frequency. Only the meta-analysis for patients not on CLB co-medication was considered, preventing any comparison with CLB-On cohort. The treatment ratio for combined CBD dose (10 + 20 mg/kg/day) vs placebo was 0.85 (P = .023). The difference over placebo did not reach statistical significance in the individual dose groups.1 The second work is a systematic review with aggregate data meta-analysis of the four RCTs, accompanied by subgroups analyses according to CLB status.2 The percentages of patients who had at least 50% reduction in seizure frequency during the treatment period were 29.1% in the CBD arm and 15.7% in the placebo group among patients not taking CLB, with a risk ratio of 1.80 (95% confidence interval 1.12-2.90; P = .015). The effect size of CBD treatment over placebo in CLB-On patients was similar to that observed in CLB-Off patients and reached statistical significance at both the 10 and 20 mg/kg/day doses. This methodology allowed evaluation of weight or contribution of each study toward the overall summary results and quantification of the degree of heterogeneity between trials and across subgroups.2 Although the responder rate was a key secondary outcome for all trials considered, its use is recommended as primary end point by the EMA’s guidelines for clinical investigations of medicines in epileptic disorders.4 Seizure response might be more appropriate than seizure reduction in evaluating the potential overall impact of CLB. Of interest, the paradoxical increases in seizure frequency observed in a subset of CLBOff patients with Lennox-Gastaut syndrome attenuated the response in change in the seizure count analysis, without affecting or having little impact on the ≥50% responder rate.3 Both studies pooled together data from trials performed in two different epileptic conditions.1,2 The RCTs overall met the criteria for meta-analytic combination due to sufficient similarity in designs, treatment protocols, method of seizure frequency assessment, and outcomes. Structured meta-analytic approach increases the statistical power to detect treatment effects within subgroups.5 Nonetheless, the lack of randomization for CLB status represents one main pitfall and may have introduced potential confounders like the degree of baseline severity and refractoriness of the subgroups. So far, this is the best available clinical evidence that CBD has antiseizure activity independent of its interaction with CLB. The open question remains whether and, mainly, to what extent CBD efficacy is enhanced by concomitant CLB administration.6,7