How-To Guides

Lemborexant showed a more favorable indirect efficacy profile than daridorexant on most subjective insomnia outcomes, while daridorexant showed a more favorable tolerability profile for some adverse events. In pooled month 1 data, lemborexant 5 mg and 10 mg had NNT values versus placebo < 10 for all reported subjective outcomes examined: ISI score < 10, ISI score ≤ 7, ISI score ≥ 6-point decrease from baseline, and subjective total sleep time (sTST) improvement > 80 minutes. In contrast, pooled daridorexant 25 mg and 50 mg had no NNT values versus placebo < 10. The authors concluded that definitive statements about relative benefit should be made cautiously because these were indirect comparisons across trials with different designs and populations.

The most robust NNT reported for lemborexant was 5. In SUNRISE 1 at month 1, both lemborexant 5 mg and 10 mg had NNT = 5 for polysomnography-defined wake after sleep onset (WASO) improvement of at least 50% from baseline, and lemborexant 10 mg also had NNT = 5 for subjective total sleep time improvement of more than 80 minutes. The article also notes that, in prior comparisons across hypnotics, month 3 NNT = 5 was reported for ISI score ≥ 6-point decrease for all doses of lemborexant and for ISI score < 10 with lemborexant 10 mg.

Yes. The article reports dose-response signals for both agents, but the pattern was clearer for lemborexant. For lemborexant, pooled month 1 NNT for sTST improvement > 80 minutes was 7 with 10 mg versus 13 with 5 mg, and the between-dose NNT was 14 (95% CI, 8-51). Somnolence with lemborexant also suggested dose response at month 1, with NNH 15 for 10 mg and 28 for 5 mg; for daridorexant, efficacy outcomes were generally numerically more robust with 50 mg than 25 mg, and NNTs for ISI score ≥ 6-point decrease from baseline were not statistically significant for 25 mg.

Daridorexant had the more favorable indirect tolerability profile for somnolence and fatigue at month 3. In pooled month 3 data, daridorexant 25 mg and 50 mg had NNH = 49 for fatigue and NNH = 85 for somnolence, whereas pooled lemborexant 5 mg and 10 mg had NNH = 29 for fatigue and NNH = 12 for somnolence. For daridorexant, the somnolence NNH was not statistically significant; for lemborexant, somnolence appeared dose dependent, with month 1 NNH 15 for 10 mg and 28 for 5 mg.

Discontinuation due to an adverse event was uncommon with both drugs, and neither pooled comparison showed a statistically significant disadvantage versus placebo. For pooled lemborexant 5 mg and 10 mg, the NNH for discontinuation due to any adverse event was 216 [NS] at month 1 and 52 [NS] at month 3. For pooled daridorexant 25 mg and 50 mg, placebo groups had higher discontinuation rates than active treatment, producing negative NNH values; the analysis therefore imputed an NNH of 1,000 for likelihood-to-be-helped-or-harmed calculations.

For both medications, LHH values were greater than 1, meaning benefit was more likely than harm for the outcomes examined. At month 3, pooled lemborexant 5 mg and 10 mg had a LHH range of 5.2-10.4 for discontinuation due to an adverse event, based on NNH = 52 [NS] and NNTs of 5-10. Pooled daridorexant 25 mg and 50 mg had a month 3 LHH range of 76.9-100 for discontinuation due to an adverse event, based on the imputed NNH of 1,000 and statistically significant NNTs of 10-13. For fatigue and somnolence specifically, month 3 LHH values were also > 1 for both agents.

For polysomnography-based sleep onset outcomes at month 1, the article describes lemborexant and daridorexant as having similar NNT ranges. PSG measures were available for cross-drug comparison only at month 1, and the indirect comparison indicated similar NNT ranges for latency to persistent sleep (LPS) improvement of at least 50% or at least 75% from baseline. By contrast, for PSG-defined WASO improvement of at least 50%, lemborexant showed more robust NNT estimates than daridorexant.

The indirect comparison is limited because the lemborexant and daridorexant trials enrolled different populations and used different designs. SUNRISE 1 for lemborexant enrolled older adults only, with women aged at least 55 years and men aged at least 65 years, required ISI ≥ 13, and did not require sleep onset difficulty; the daridorexant studies enrolled adults aged at least 18 years, required ISI ≥ 15, and were limited to patients with a history of subjective sleep onset latency at least 30 minutes. Mean baseline LPS was shorter in SUNRISE 1 (44.6-44.9 minutes) than in the daridorexant studies (63.6-71.8 minutes), and the trials also differed in ethnic diversity, duration, and availability of PSG measures. Because of this heterogeneity, the authors caution against making definitive comparative claims without direct head-to-head trials.

No. The study supports both drugs as treatment options for adults with insomnia, but it does not prove superiority because the comparison was indirect. The authors conclude that lemborexant had a more favorable NNT efficacy profile, while daridorexant had more favorable NNH estimates for discontinuation due to an adverse event and a tolerability advantage for somnolence. They also state that definitive statements on relative benefits should be made with caution and that direct head-to-head DORA trials are needed.