Lunesta vs Dayvigo: Switching Between Them

How These Two Drugs Work

Eszopiclone and lemborexant both treat insomnia, but they act on entirely different molecular targets. Eszopiclone binds to the benzodiazepine site on GABA-A receptors, increasing chloride channel conductance and producing sedation through broad CNS depression. Lemborexant blocks orexin-1 and orexin-2 receptors, preventing the wake-promoting neuropeptides orexin-A and orexin-B from sustaining wakefulness.

Eszopiclone: GABA-A Modulation

Eszopiclone is the S-enantiomer of zopiclone. The FDA approved it in 2004 under the brand name Lunesta for the treatment of insomnia characterized by difficulty with sleep onset or sleep maintenance. Its primary mechanism produces sedation, anxiolysis, and muscle relaxation, which explains both its efficacy and its side-effect burden. Prescribing information is available on the FDA database.

Lemborexant: Orexin Receptor Antagonism

Lemborexant received FDA approval in December 2019. Rather than suppressing the entire CNS, it selectively removes the orexin-driven "wake signal," theoretically allowing the brain's natural sleep circuitry to take over. This mechanistic difference is why SUNRISE-1 investigators reported favorable next-morning psychomotor performance data compared with zolpidem extended-release. Suvorexant (Belsomra) shares this DORA mechanism; lemborexant has a shorter half-life of roughly 17-19 hours versus suvorexant's 12 hours, though individual variability applies.

Efficacy: What the Trial Data Show

No randomized, double-blind, head-to-head trial has directly compared eszopiclone with lemborexant. The following analysis draws on the best available single-arm and active-comparator data from each drug's key program.

Lunesta: Krystal 2003 Six-Month Data

Krystal et al. Published a six-month, double-blind, placebo-controlled trial of eszopiclone 3 mg in adults with chronic insomnia. That trial (N=788) showed statistically significant reductions in subjective sleep-onset latency, wake time after sleep onset, and number of awakenings throughout all six months. Mean sleep time increased by approximately 37 minutes over placebo. Critically, no tolerance to the sleep-maintenance benefit was detected across the 26-week observation window, which was a concern with earlier z-drugs. The FDA label for eszopiclone reflects this durable efficacy finding, noting no evidence of tolerance through six months.

The 1 mg and 2 mg doses were approved later for patients at higher risk of next-day impairment, including women and older adults, after pharmacokinetic data showed that the 3 mg dose produced plasma concentrations sufficient to impair driving for up to 11 hours post-dose in some individuals.

Dayvigo: SUNRISE-1 and SUNRISE-2

SUNRISE-1 was a phase 3, randomized, double-blind, placebo- and active-controlled trial published in JAMA Network Open in 2019. The trial (N=291) compared lemborexant 5 mg and 10 mg against zolpidem extended-release 6.25 mg and placebo over one month. Lemborexant 10 mg reduced subjective sleep-onset latency by 22.0 minutes versus 14.0 minutes for zolpidem ER, and wake after sleep onset by 40.5 minutes versus 32.0 minutes for zolpidem ER. Both lemborexant doses outperformed placebo on polysomnographic endpoints. The lemborexant 5 mg arm showed numerically better next-morning driving performance than zolpidem ER on day 9 of treatment.

SUNRISE-2 extended the observation period to 12 months. That study (N=949) confirmed sustained efficacy for lemborexant 5 mg and 10 mg on sleep onset and maintenance without tolerance development. This mirrors the no-tolerance finding from the Krystal eszopiclone trial, suggesting both agents maintain benefit over months of use.

Side-by-Side Efficacy Summary

Sleep-onset latency reduction, sleep maintenance improvement, and absence of tolerance are documented for both drugs. Direct numerical comparison across trials is complicated by different study populations, different PSG protocols, and different active comparators. The American Academy of Sleep Medicine 2017 clinical practice guideline for chronic insomnia assigned a "weak" recommendation for both eszopiclone and lemborexant class agents, reflecting adequate efficacy evidence offset by tolerability concerns specific to each mechanism.

Safety and Side Effects

The safety profiles diverge substantially because the mechanisms are different.

Eszopiclone Safety Concerns

The most commonly reported adverse effect with eszopiclone is an unpleasant metallic or bitter taste, occurring in approximately 34% of patients at the 3 mg dose in the Krystal trial. The FDA updated the eszopiclone label in 2014 to lower the recommended starting dose to 1 mg for all adults after post-marketing data confirmed next-morning impairment at higher doses. Complex sleep behaviors, including sleep-driving and sleep-eating, are listed as class-wide risks for all sedative-hypnotics under the FDA's 2019 boxed warning. The FDA's 2019 safety communication on complex sleep behaviors covers eszopiclone explicitly.

Dependence potential is real. As a Schedule IV substance, eszopiclone carries abuse liability. Abrupt discontinuation after prolonged use at 3 mg may produce rebound insomnia and, in some patients, withdrawal symptoms including anxiety and tremor. A slow taper is the standard approach when stopping.

Lemborexant Safety Concerns

Lemborexant's most common adverse effect in SUNRISE-1 was somnolence, reported in 10% of the 10 mg group versus 1% of placebo. The FDA label for lemborexant includes a warning about next-morning impairment, sleep paralysis, hypnagogic hallucinations, and cataplexy-like symptoms, all of which relate to orexin suppression. The complex sleep behavior boxed warning applies here as well. In patients with narcolepsy or cataplexy, lemborexant is contraindicated because orexin signaling is already deficient.

Lemborexant is metabolized primarily by CYP3A. Co-administration with strong CYP3A inhibitors (e.g., itraconazole, clarithromycin) raises lemborexant plasma levels substantially; the dose should not exceed 5 mg in that setting. CYP3A inducers (e.g., rifampin, carbamazepine) reduce exposure and may compromise efficacy.

Respiratory Safety

This distinction matters clinically. Eszopiclone, as a GABA-A modulator, carries respiratory depression risk and should be used with caution in patients with moderate-to-severe obstructive sleep apnea (OSA) or COPD. A 2019 review in the Annals of the American Thoracic Society summarized evidence that non-benzodiazepine hypnotics suppress respiratory drive. Lemborexant, by contrast, does not appear to worsen respiratory drive in OSA patients. A dedicated phase 1 study in mild-to-moderate OSA patients found no clinically meaningful change in apnea-hypopnea index with lemborexant 10 mg, though caution in severe OSA is still warranted pending larger trials.

Who Is Each Drug Better For?

Neither drug is universally superior. Patient-specific factors determine which agent fits better.

Patients Who May Do Better on Eszopiclone

• Adults without respiratory compromise who need both sleep-onset and sleep-maintenance coverage.
• Patients who have previously failed benzodiazepine-class agents and want a non-benzodiazepine GABA option.
• Individuals for whom cost is a primary concern: generic eszopiclone is widely available and costs significantly less than branded Dayvigo, often under $30 for a 30-day supply with a GoodRx coupon versus $300-plus for Dayvigo without insurance.
• Short-term use of 2-4 weeks where dependence risk is lower.

Patients with a history of substance use disorder should approach eszopiclone cautiously given its Schedule IV abuse potential and the documented risk of non-medical use of z-drugs reported in SAMHSA surveillance data.

Patients Who May Do Better on Lemborexant

• Adults with mild-to-moderate OSA where respiratory safety of a GABA agent is a concern.
• Patients who develop tolerance or dependence on z-drugs and need to transition off eszopiclone.
• Older adults: the 5 mg dose showed a numerically favorable next-morning impairment profile in the SUNRISE-1 driving simulation substudy, which matters for fall risk in patients over 65. The American Geriatrics Society Beers Criteria 2023 update lists non-benzodiazepine hypnotics including eszopiclone as potentially inappropriate in older adults.
• Patients who find the metallic taste of eszopiclone intolerable. This is a common and underappreciated reason for non-adherence.

Switching from Lunesta to Dayvigo

Switching is clinically reasonable and fairly common. There is no published randomized trial specifically studying the optimal switch protocol, so the following guidance is based on pharmacokinetic principles, FDA labeling, and standard clinical practice.

Why Patients Switch

The most common documented reasons for switching from eszopiclone to lemborexant are persistent metallic taste, next-morning grogginess, concern about dependence, a new diagnosis of OSA making GABA-modulator use less desirable, or inadequate sleep maintenance despite dose optimization.

The Cross-Taper Approach

Abrupt discontinuation of eszopiclone after weeks or months of daily use at 2-3 mg risks rebound insomnia. A cross-taper over 2-4 weeks reduces this risk. A reasonable clinical approach:

1. Week 1-2: Start lemborexant 5 mg on nights when the patient would otherwise take eszopiclone. On alternating nights, maintain the current eszopiclone dose.
2. Week 3: Take lemborexant every night. Reduce eszopiclone to every-other-night or drop to 1 mg nightly.
3. Week 4: Discontinue eszopiclone entirely. Titrate lemborexant to 10 mg if 5 mg is insufficient.

This schedule is individualized; patients with longer duration of eszopiclone use or higher doses may need a slower taper. The FDA eszopiclone prescribing information advises tapering rather than abrupt cessation but does not specify a fixed schedule.

Monitoring During the Switch

Patients should track subjective sleep quality, next-morning alertness, and any withdrawal-type symptoms (anxiety, irritability, increased dreaming) during the taper. If rebound insomnia is severe during eszopiclone reduction, slowing the taper is appropriate. Cognitive behavioral therapy for insomnia (CBT-I) has Level A evidence as a first-line treatment in AASM guidelines and can be started concurrently during the switch to improve long-term outcomes without adding pharmacologic burden.

Switching from Dayvigo to Lunesta

The reverse switch is less common but occasionally requested. Lemborexant does not produce physical dependence in the same way as GABA-modulators, so abrupt discontinuation is generally tolerable. However, patients may experience several nights of slightly worsened sleep onset as the orexin system readjusts. Starting eszopiclone at 1 mg on the first night of discontinuation and titrating as needed over one to two weeks is a reasonable approach. Confirm there are no CYP3A-related drug interactions before initiating eszopiclone alongside any concurrent medications.

Drug Interactions and Special Populations

CYP Interactions

Eszopiclone is metabolized by CYP3A4 and CYP2E1. Strong CYP3A4 inhibitors (ketoconazole, clarithromycin) increase eszopiclone exposure; the dose should not exceed 2 mg when these are co-prescribed. The eszopiclone label specifically warns about this interaction. Alcohol potentiates CNS depression with eszopiclone and is contraindicated with both agents. Lemborexant shares the CYP3A4 sensitivity, as noted above.

Concurrent use of opioids with either agent increases respiratory depression risk. The FDA's 2016 boxed warning on opioid-benzodiazepine combinations was later extended conceptually to all CNS depressants, and clinicians should exercise the same caution when combining opioids with eszopiclone or lemborexant.

Pregnancy and Lactation

Neither drug has adequate human pregnancy safety data. Both are categorized as risks to the fetus based on animal data. The NIH LactMed database notes insufficient data on eszopiclone transfer into breast milk. Lemborexant data in lactation are similarly limited. Non-pharmacologic interventions, particularly CBT-I, should be the default for pregnant and breastfeeding patients with insomnia.

Hepatic Impairment

Severe hepatic impairment reduces clearance of both drugs. Eszopiclone should be limited to 2 mg in severe hepatic impairment. Lemborexant should not exceed 5 mg in moderate hepatic impairment and is not recommended in severe hepatic impairment per FDA labeling.

Dosing Quick Reference

| Parameter | Eszopiclone (Lunesta) | Lemborexant (Dayvigo) | |---|---|---| | Starting dose (adults) | 1 mg at bedtime | 5 mg at bedtime | | Maximum dose | 3 mg | 10 mg | | Recommended dose (women / elderly) | 1 mg | 5 mg | | Time to peak plasma (Tmax) | ~1 hour | ~1-3 hours | | Half-life | ~6 hours | ~17-19 hours | | Schedule | IV | IV | | Generic available | Yes | No (as of 2025) | | Approved duration | Chronic use (no stated limit) | Chronic use (no stated limit) |

Cost and Access

Generic eszopiclone launched in 2014 after the Sepracor patent expired. A 30-day supply of eszopiclone 2 mg typically costs $20-40 at major pharmacies with discount programs. Lemborexant remains branded through at least 2025; retail cost is approximately $300-400 per month without insurance. Most commercial insurance plans and Medicare Part D cover both agents, though lemborexant frequently requires prior authorization. Patients switching to lemborexant for medical reasons (e.g., OSA diagnosis, documented adverse effects on eszopiclone) generally have higher prior-authorization approval rates when the clinical rationale is documented.