What was the exact primary endpoint result for lemborexant 10 mg in SUNRISE-1?

Lemborexant 10 mg reduced latency to persistent sleep by 10.5 minutes more than placebo at nights 29/30 (95% CI: −16.3 to −4.7 minutes, p <0.001). The absolute LPS at study end was 17.5 minutes, down from a baseline of 43.4 minutes.

How quickly does lemborexant start working based on SUNRISE-1 data?

The 10 mg dose showed significant separation from placebo on the very first night of dosing (−12.2 minutes, p <0.001). The 5 mg dose also reached significance at nights 1/2 but with a smaller effect that grew to near-maximum by week 2.

Did lemborexant beat zolpidem in SUNRISE-1?

The trial was not designed or powered for superiority testing against zolpidem ER 6.25 mg. Descriptive comparisons show lemborexant 10 mg was numerically superior on most PSG endpoints, particularly WASO, but formal statistical conclusions cannot be drawn.

What age group was studied in SUNRISE-1?

SUNRISE-1 enrolled adults aged 55 years and older. The subsequent SUNRISE-2 trial expanded enrollment to adults 18 years and older, broadening the evidence base to younger insomnia patients.

How large was the placebo response in SUNRISE-1?

Placebo-treated patients improved by approximately 16 minutes in LPS over 30 nights. This substantial placebo effect is typical in PSG-based insomnia trials and reflects regression to the mean, adaptation to the sleep lab environment, and nonspecific therapeutic contact effects.

Does lemborexant help with staying asleep or just falling asleep?

Both. SUNRISE-1 demonstrated significant reductions in WASO (wake after sleep onset) of 16-17 minutes versus placebo, in addition to the sleep onset benefits. This dual-domain efficacy reflects orexin receptor antagonism's effect on both sleep initiation and maintenance pathways.

Was there any tolerance development over 30 days in SUNRISE-1?

No. The treatment effect remained stable from the first assessment through night 29/30 for both doses. There was no evidence of diminishing efficacy (tachyphylaxis) within the study period. Longer-term data from SUNRISE-2 confirmed persistence through 12 months.

What is the clinical significance of a 10-minute reduction in sleep onset latency?

A 10-minute PSG-measured reduction is considered clinically meaningful in the regulatory context. It brings mean sleep onset from the pathological range (above 40 minutes) closer to the normal range (under 20 minutes). Combined with WASO improvements, total additional sleep time gained is approximately 25-30 minutes nightly.

How does SUNRISE-1 support the FDA-approved dosing of Dayvigo?

SUNRISE-1 demonstrated efficacy for both the 5 mg and 10 mg doses, with the FDA approving a 5 mg recommended starting dose and 10 mg maximum dose. The incremental benefit of 10 mg over 5 mg was modest for LPS but present across all endpoints.

Were there any subgroups that did not respond to lemborexant in SUNRISE-1?

No pre-specified subgroup showed a qualitative treatment interaction (no subgroup was harmed). Treatment effects were consistent across age strata, sex, geographic region, and baseline insomnia severity, though individual variability within all subgroups was high.

SUNRISE-1 Results in Detail: Numbers, Subgroups, and Time Course

Q: How quickly does lemborexant start working based on SUNRISE-1 data?

The 10 mg dose showed significant separation from placebo on the very first night of dosing (−12.2 minutes, p <0.001). The 5 mg dose also reached significance at nights 1/2 but with a smaller effect that grew to near-maximum by week 2.

Q: Did lemborexant beat zolpidem in SUNRISE-1?

The trial was not designed or powered for superiority testing against zolpidem ER 6.25 mg. Descriptive comparisons show lemborexant 10 mg was numerically superior on most PSG endpoints, particularly WASO, but formal statistical conclusions cannot be drawn.

Q: What age group was studied in SUNRISE-1?

SUNRISE-1 enrolled adults aged 55 years and older. The subsequent SUNRISE-2 trial expanded enrollment to adults 18 years and older, broadening the evidence base to younger insomnia patients.

Q: How large was the placebo response in SUNRISE-1?

Placebo-treated patients improved by approximately 16 minutes in LPS over 30 nights. This substantial placebo effect is typical in PSG-based insomnia trials and reflects regression to the mean, adaptation to the sleep lab environment, and nonspecific therapeutic contact effects.

Q: Does lemborexant help with staying asleep or just falling asleep?

Both. SUNRISE-1 demonstrated significant reductions in WASO (wake after sleep onset) of 16-17 minutes versus placebo, in addition to the sleep onset benefits. This dual-domain efficacy reflects orexin receptor antagonism's effect on both sleep initiation and maintenance pathways.

Q: Was there any tolerance development over 30 days in SUNRISE-1?

No. The treatment effect remained stable from the first assessment through night 29/30 for both doses. There was no evidence of diminishing efficacy (tachyphylaxis) within the study period. Longer-term data from SUNRISE-2 confirmed persistence through 12 months.

Q: What is the clinical significance of a 10-minute reduction in sleep onset latency?

A 10-minute PSG-measured reduction is considered clinically meaningful in the regulatory context. It brings mean sleep onset from the pathological range (above 40 minutes) closer to the normal range (under 20 minutes). Combined with WASO improvements, total additional sleep time gained is approximately 25-30 minutes nightly.

Q: How does SUNRISE-1 support the FDA-approved dosing of Dayvigo?

SUNRISE-1 demonstrated efficacy for both the 5 mg and 10 mg doses, with the FDA approving a 5 mg recommended starting dose and 10 mg maximum dose. The incremental benefit of 10 mg over 5 mg was modest for LPS but present across all endpoints.

Q: Were there any subgroups that did not respond to lemborexant in SUNRISE-1?

No pre-specified subgroup showed a qualitative treatment interaction (no subgroup was harmed). Treatment effects were consistent across age strata, sex, geographic region, and baseline insomnia severity, though individual variability within all subgroups was high.

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |-----------|--------| | N | 1,006 randomized (modified ITT: 949) | | Intervention | Lemborexant 5 mg (LEM5) and 10 mg (LEM10) nightly | | Comparator | Placebo and zolpidem ER 6.25 mg (active reference) | | Duration | 30 nights (with 2-night PSG assessments at baseline, night 1/2, night 15/16, night 29/30) | | Primary endpoint | Change from baseline in latency to persistent sleep (LPS) at nights 29/30 vs placebo | | Key result | LEM5: −7.6 min vs placebo (p = 0.019); LEM10: −10.5 min vs placebo (p <0.001) |

Trial Design and Measurement Approach

SUNRISE-1 was a Phase 3, randomized, double-blind, parallel-group study conducted across 66 sites in North America, Europe, and Asia. The trial enrolled adults aged 55 years and older with insomnia disorder per DSM-5 criteria. Participants required objective confirmation: a baseline latency to persistent sleep (LPS) of 25 minutes or greater on at least one screening PSG night.

The four-arm design (placebo, LEM5, LEM10, zolpidem ER 6.25 mg) randomized in a 2:2:2:1 ratio. This asymmetric allocation gave each lemborexant arm adequate power against placebo while limiting zolpidem exposure since it served as a reference, not a formal comparator for superiority testing.

PSG recordings occurred across two consecutive nights at three timepoints. The investigators averaged values across each two-night window. This averaging approach reduces night-to-night variability inherent in single-night PSG, a methodological strength that strengthens the signal-to-noise ratio.

Primary Endpoint: Latency to Persistent Sleep

The primary analysis used a mixed-effects model for repeated measures (MMRM) with treatment, visit, treatment-by-visit interaction, region, and baseline LPS as covariates. Results at the primary timepoint (nights 29/30):

| Arm | Baseline LPS (min) | Night 29/30 LPS (min) | LS Mean Change | Difference vs Placebo (95% CI) | p-value | |-----|--------------------|-----------------------|----------------|--------------------------------|---------| | Placebo (n = 135) | 44.5 | 28.4 | −16.0 |, |, | | LEM5 (n = 266) | 44.7 | 20.4 | −23.6 | −7.6 (−13.4 to −1.8) | 0.019 | | LEM10 (n = 269) | 43.4 | 17.5 | −26.5 | −10.5 (−16.3 to −4.7) | <0.001 | | Zolpidem ER 6.25 mg (n = 134) | 42.0 | 20.9 | −22.4 | −6.4 (−13.0 to 0.2) |, |

The placebo response was substantial (16 minutes of improvement), consistent with known first-night effects in PSG trials and regression to the mean in populations selected for elevated LPS. Despite this, both lemborexant doses cleared the pre-specified gatekeeping procedure.

Time-Course Pattern: When Treatment Separation Emerged

A critical question for clinicians is how quickly the drug works. SUNRISE-1's repeated PSG assessments allow tracking of the treatment effect over time:

| Timepoint | LEM5 vs Placebo (min) | LEM10 vs Placebo (min) | |-----------|-----------------------|------------------------| | Nights 1/2 | −6.7 (p = 0.021) | −12.2 (p <0.001) | | Nights 15/16 | −8.0 (p = 0.005) | −10.0 (p <0.001) | | Nights 29/30 | −7.6 (p = 0.019) | −10.5 (p <0.001) |

LEM10 produced its largest placebo-adjusted effect on the very first nights of dosing, with the magnitude remaining stable through day 30. LEM5 showed a more gradual onset, reaching near-maximal effect by week 2. Neither dose showed tachyphylaxis over the study period. This stable effect trajectory distinguishes lemborexant from some older hypnotics where tolerance development can erode efficacy within weeks.

Secondary Endpoints: Sleep Maintenance and Efficiency

The secondary endpoint hierarchy tested sleep efficiency (SE) and wake after sleep onset (WASO) at nights 29/30:

Sleep Efficiency

| Arm | Baseline SE (%) | Night 29/30 SE (%) | LS Mean Change | Difference vs Placebo (95% CI) | |-----|----------------|--------------------|-----------------|---------------------------------| | Placebo | 67.3 | 75.1 | +7.8 |, | | LEM5 | 66.5 | 79.9 | +13.4 | +5.6 (3.0 to 8.2), p <0.001 | | LEM10 | 67.4 | 81.3 | +13.9 | +6.1 (3.5 to 8.7), p <0.001 |

Wake After Sleep Onset

| Arm | Baseline WASO (min) | Night 29/30 WASO (min) | LS Mean Change | Difference vs Placebo (95% CI) | |-----|---------------------|------------------------|----------------|----------------------------------| | Placebo | 88.8 | 66.2 | −22.6 |, | | LEM5 | 92.0 | 53.4 | −38.6 | −16.0 (−24.4 to −7.6), p <0.001 | | LEM10 | 86.9 | 47.4 | −39.5 | −16.9 (−25.2 to −8.5), p <0.001 |

The WASO reductions are clinically meaningful. A 16-17 minute reduction in nighttime wakefulness translates to patients spending roughly 20 fewer minutes lying awake in the second half of the night. For a population with baseline WASO approaching 90 minutes, this represents an 18-19% relative improvement.

Zolpidem ER Comparisons (Descriptive Only)

The trial was not powered for direct superiority testing against zolpidem ER 6.25 mg. However, the descriptive data are informative:

LPS: Zolpidem ER showed a numerically similar reduction to LEM5, with both trailing LEM10.
WASO: Both lemborexant doses produced numerically greater WASO reductions than zolpidem ER (difference approximately 8-10 minutes). This aligns with the pharmacologic rationale: orexin blockade addresses sleep maintenance through a different mechanism than GABA-A modulation.
SE: LEM10 produced the highest sleep efficiency of any arm.

The FDA prescribing information for Dayvigo references SUNRISE-1 as a key efficacy study supporting the 5 mg and 10 mg approved doses.

Response Distribution and Variability

Individual-level response data were not fully published in the primary manuscript, though several indicators of response heterogeneity are available:

Standard deviations for LPS change scores were large (approximately 25-30 minutes across arms), indicating substantial inter-individual variability. Some patients achieved near-normal sleep onset (<10 minutes), while others showed minimal response.

The modified ITT population excluded 57 of 1,006 randomized patients (5.7%), primarily due to missing PSG data rather than treatment failure. Sensitivity analyses using pattern-mixture models for missing data produced consistent results, suggesting dropout patterns did not meaningfully bias the primary estimates.

Subgroup Analyses

Pre-specified subgroup analyses examined consistency of treatment effects across:

Age: Patients 65 years and older showed treatment effects of similar or slightly greater magnitude compared to the 55-64 cohort. This is relevant because the American Academy of Sleep Medicine guidelines note particular concern about hypnotic safety in older adults.
Sex: Effects were consistent between male and female participants.
Region: North American, European, and Asian sites produced broadly comparable results, though sample sizes within regions limited precision.
Baseline severity: Patients with higher baseline LPS (more severe sleep onset difficulty) tended to show larger absolute reductions, though relative improvements were similar across severity strata.

No subgroup showed a qualitative interaction (treatment harm), supporting generalizability of the findings within the studied population.

Limitations the Authors Acknowledged

The investigators noted several constraints. The population was restricted to adults 55 years and older, limiting direct extrapolation to younger insomnia patients (addressed later in the SUNRISE-2 trial, which included adults 18+). The 30-night duration, while adequate for regulatory endpoints, does not address long-term efficacy maintenance. The zolpidem ER arm used the lower 6.25 mg dose approved for older adults, not the 12.5 mg dose used in younger populations.

PSG-measured endpoints, while objective, do not fully capture the patient experience of insomnia. Subjective sleep measures (patient-reported sleep quality, next-day functioning) were collected as exploratory endpoints and generally supported the PSG findings, though effect sizes were smaller on subjective measures, a pattern common across insomnia pharmacotherapy trials.

The SUNRISE-2 trial subsequently addressed duration limitations by evaluating lemborexant over 12 months using subjective endpoints, confirming sustained efficacy without evidence of rebound insomnia upon discontinuation.

Clinical Translation

For prescribers considering lemborexant, SUNRISE-1's results indicate:

The 10 mg dose provides approximately 3 additional minutes of LPS reduction versus 5 mg, with comparable WASO benefits. Given the relatively modest incremental benefit on sleep onset, the 5 mg starting dose recommended in the Dayvigo label is appropriate for most patients, with uptitration to 10 mg reserved for inadequate responders.

The dual benefit on both sleep onset and sleep maintenance distinguishes DORAs from traditional hypnotics that primarily target one domain. For patients whose insomnia involves both difficulty falling asleep and nighttime awakenings, this dual-action profile may reduce the need for combination approaches.

Frequently asked questions

›

References

Rosenberg R, Murphy P, Zammit G, et al. Comparison of lemborexant with placebo and zolpidem tartrate extended release for the treatment of older adults with insomnia disorder: a phase 3 randomized clinical trial. JAMA Netw Open. 2019;2(12):e1918254. https://pubmed.ncbi.nlm.nih.gov/31886325/
Kärppä M, Yardley J, Pinner K, et al. Long-term efficacy and tolerability of lemborexant compared with placebo in adults with insomnia disorder: results from the phase 3 randomized clinical trial SUNRISE-2. Sleep. 2020;43(9):zsaa123. https://pubmed.ncbi.nlm.nih.gov/32621681/
U.S. Food and Drug Administration. Dayvigo (lemborexant) prescribing information. 2019. https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/212028s000lbl.pdf
Sateia MJ, Buysse DJ, Krystal AD, Neubauer DN, Heald JL. Clinical practice guideline for the pharmacologic treatment of chronic insomnia in adults: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2017;13(2):307-349. https://pubmed.ncbi.nlm.nih.gov/28942757/
Herring WJ, Connor KM, Ivgy-May N, et al. Suvorexant in patients with insomnia: results from two 3-month randomized controlled clinical trials. Biol Psychiatry. 2016;79(2):136-148. https://pubmed.ncbi.nlm.nih.gov/25526970/