Dayvigo Switching Protocols: How to Switch From or To Lemborexant Safely

Lemborexant, sold under the brand name Dayvigo, is a dual orexin receptor antagonist (DORA) approved by the FDA at 5 mg and 10 mg oral tablet doses for adults with insomnia. It is a DEA Schedule IV controlled substance. It is not the same drug as suvorexant (Belsomra), a related but pharmacokinetically distinct DORA, and it is not a benzodiazepine, a z-drug, or a melatonin receptor agonist, even though patients often switch to it from one of those classes.

The direct answer: switching to or from lemborexant is a timing and overlap problem, not a same-night substitution. Lemborexant's active metabolite (M4) has a half-life of roughly 40 hours, far longer than the parent drug's 17 to 19 hours, so any prior CNS depressant still in the body when lemborexant starts, or still present after lemborexant stops, adds to sedation risk for longer than a single overnight window. The FDA label does not provide a specific cross-taper schedule for switching between hypnotics; the taper sequences described below are clinical practice extrapolations, not an FDA-directed protocol, and should be individualized with a prescriber.

Evidence boundary: what is established, what is not

Established from the FDA label and trial evidence: lemborexant's receptor mechanism, its approved dose range, its metabolic pathway through CYP3A4, its contraindication in narcolepsy, and its general safety profile relative to placebo in the pivotal trials.

Plausible but not directly tested: the specific taper sequences for zolpidem, eszopiclone, or benzodiazepines described in this article. No published randomized trial has tested a structured cross-taper protocol from another hypnotic onto lemborexant. The sequences here reflect pharmacokinetic reasoning and general sedative-taper principles, not a dedicated switching trial.

Not established: direct dose equivalence between lemborexant and suvorexant, or between lemborexant and any benzodiazepine or z-drug. No head-to-head trial comparing lemborexant to suvorexant on sleep architecture endpoints appears among the sources reviewed for this article. Treat any stated equivalence between these drugs as unverified until a prescriber confirms it against current evidence.

How lemborexant works

Lemborexant blocks orexin-1 (OX1R) and orexin-2 (OX2R) receptors in the lateral hypothalamus, preventing orexin-A and orexin-B neuropeptides from sustaining the wake-promoting signaling that keeps a person alert into the sleep window [2]. Orexin neurons project to the locus coeruleus, raphe nuclei, tuberomammillary nucleus, and basal forebrain, and this signaling is thought to be dysregulated in chronic insomnia [2].

This is a narrower mechanism than the one used by benzodiazepines and z-drugs (zolpidem, eszopiclone, zaleplon), which enhance GABA-A receptor activity broadly across the brain. Suvorexant, the first DORA approved by the FDA, binds the same two receptors but with roughly two-fold higher affinity for OX2R and different binding kinetics than lemborexant [3]. Because the two drugs are not identical at the receptor level, they should not be assumed dose-equivalent.

What the pivotal trials showed, and what still needs verification

Lemborexant's approval rested primarily on SUNRISE-1 and SUNRISE-2, two phase 3 randomized, double-blind trials published in JAMA Network Open and Sleep, respectively [1, 6]. In general terms, these trials found that lemborexant reduced subjective and polysomnographic measures of sleep onset latency and wake after sleep onset compared with placebo, with effects that persisted through longer-term follow-up in SUNRISE-2 without evidence of rebound insomnia at discontinuation [6].

The same caution applies to the next-morning driving simulator data. The prior draft attributed a specific driving-impairment comparison (lemborexant versus zopiclone, measured by standard deviation of lateral position) to a paper indexed here as PMID 31252205, but that PMID corresponds to a lemborexant study in irregular sleep-wake rhythm disorder, not a driving simulator study in insomnia. The FDA label does address next-morning psychomotor and driving performance findings from the development program [5], and that label is the safer primary source to cite for this claim until the correct trial publication is identified and linked.

Pharmacokinetics that drive switching decisions

Lemborexant reaches peak plasma concentration about 1 to 3 hours after an oral dose; taking it with a heavy meal delays peak concentration and slightly lowers it, so the label-based guidance is to take it on an empty stomach or after a light meal [5]. It is metabolized primarily by CYP3A4. Strong CYP3A4 inhibitors (clarithromycin, itraconazole, ritonavir) can raise exposure several-fold and are not recommended for co-administration; moderate inhibitors require capping the dose at 5 mg. Strong CYP3A4 inducers (rifampin, carbamazepine, phenytoin) can substantially lower exposure, and the label advises avoiding lemborexant with these agents [5].

The parent drug's 17 to 19 hour half-life and the M4 metabolite's roughly 40 hour half-life mean that even after a single dose, measurable drug remains in the body well past the 7 to 9 hour sleep window. This is the pharmacologic reason overlap with another sedating drug should be brief and supervised, not the basis for a specific numeric washout rule.

Should you stop the old sleep medication before starting lemborexant, or overlap doses?

There is no FDA-approved switching protocol answering this question directly. The general principle used in clinical practice is to avoid stacking two full-strength sedating drugs on the same night, while also avoiding an abrupt stop of a drug that causes physical dependence.

From zolpidem. Zolpidem is a GABA-A positive allosteric modulator, and stopping it abruptly after more than about four weeks of nightly use can cause rebound insomnia, anxiety, tremor, and rarely seizures, especially at higher doses [7]. A commonly used clinical approach is to reduce the zolpidem dose gradually over two to four weeks while introducing lemborexant at 5 mg, rather than stopping zolpidem outright on the night lemborexant starts. This is a clinical judgment call informed by general sedative-taper principles [7], not a tested protocol, and the schedule needs to be individualized by the prescribing clinician based on dose, duration of use, and the patient's medical history.

From eszopiclone. Eszopiclone's half-life is around 6 hours in healthy adults and longer in older adults; the FDA has lowered its recommended starting dose because of next-morning impairment concerns, particularly in women and older adults. A gradual step-down of eszopiclone with lemborexant 5 mg introduced near the end of the taper follows the same overlap-avoidance logic used for zolpidem, again without a dedicated switching trial to confirm timing.

From suvorexant. Because both drugs occupy the same receptors, the main concern is pharmacodynamic overlap rather than withdrawal. Suvorexant's half-life is roughly 12 hours, shorter than lemborexant's M4 metabolite half-life [3, 5]. A short washout of a night or two after the last suvorexant dose, before starting lemborexant, reduces the chance of additive orexin blockade during the switch, though patients should expect one or two nights of disrupted sleep during that gap. The reverse switch (lemborexant to suvorexant) follows the same logic in the opposite direction.

From benzodiazepines. This is the highest-risk switch. Physical dependence on a benzodiazepine can develop within four to six weeks of nightly therapeutic-dose use, and abrupt discontinuation after long-term use carries a real seizure risk that lemborexant does not treat or prevent. Clinical guidelines generally favor tapering the benzodiazepine first, using gradual dose reductions over an extended period, before or during the introduction of a non-benzodiazepine option [7]. A commonly cited approach in clinical practice (sometimes referenced as the Ashton Manual method) uses small, slow dose reductions over weeks to months using a long-acting benzodiazepine equivalent; this specific percentage-based schedule is widely used in practice but is not tied to a primary source in this article's reference list, and its details should be verified independently before being presented to a patient as a fixed rule. Introducing lemborexant, at 5 mg only, is generally reserved until the benzodiazepine dose is well below its starting point and withdrawal symptoms are stable.

From ramelteon. Ramelteon is a melatonin receptor agonist with no known abuse potential and no GABA-A or orexin activity, so no pharmacodynamic overlap with lemborexant is expected. A meta-analysis found a modest sleep-onset benefit for ramelteon over placebo [11]. Patients switching because of inadequate response can generally stop ramelteon the night lemborexant starts, without a taper.

When 10 mg instead of 5 mg is appropriate

The approved starting dose is 5 mg. The label allows escalation to 10 mg if 5 mg is tolerated but insufficiently effective, and escalation is reasonable no sooner than about a week at the lower dose [5]. If a patient reports grogginess persisting past 7 to 8 hours after a dose, the 10 mg dose should not be used. In the trial evidence reviewed for this article, the 10 mg dose produced a numerically greater improvement in sleep onset measures than 5 mg, though the exact size of that difference needs the primary-source verification noted above before it is quoted as a specific figure.

Special populations

Older adults (65 and older). The label reports higher drug exposure in this group compared with younger adults, and recommends starting at 5 mg [5]. The 2023 American Geriatrics Society Beers Criteria flags orexin receptor antagonists as a class for fall-risk review before prescribing in older adults [12]. A falls assessment before starting, and after any dose increase, is reasonable practice.

Hepatic impairment. Moderate impairment increases exposure and caps the recommended dose at 5 mg; severe hepatic impairment (Child-Pugh C) is a contraindication under the label [5].

Substance use disorder history. Lemborexant and suvorexant are both Schedule IV. An abstract-level report comparing lemborexant to triazolam in recreational drug users described lower drug-liking scores for lemborexant, suggesting reduced but not absent abuse potential [13]; because this is conference-abstract evidence rather than a peer-reviewed full report, it carries more uncertainty than the phase 3 trial data, and the correct citation for this specific abstract should be confirmed before publication. Controlled-substance monitoring is still appropriate when switching a patient with this history from a z-drug to a DORA.

Clinician conversation and monitoring framework for a lemborexant switch

This framework is a structured way to have the switching conversation and track it over time. It distinguishes what the FDA label actually specifies from what depends on individual clinical judgment.

The rows above marked "requires individualized judgment" are the parts of a switch where no clinical trial has directly tested the schedule. A patient or caregiver should treat these as discussion points with the prescriber rather than fixed rules.

When to seek urgent care instead of adjusting the dose

Sudden difficulty breathing, unresponsiveness, a seizure, or a fall with injury during a switch is an emergency and needs immediate medical attention rather than a change in the taper schedule. Rebound insomnia and mild anxiety during a benzodiazepine or z-drug taper are uncomfortable but expected; a seizure or marked confusion is not, and should prompt same-day contact with the prescriber or emergency care.

Editorial and delivery notes: This draft flags four verification items for the reviewing clinician before publication: (1) the SUNRISE-1 trial design description (sample size, arm structure, comparator, and exact efficacy figures) needs to be pulled from the primary publication rather than reused from the prior draft; (2) the next-morning driving-simulator claim needs a correctly matched primary source, since the previously cited PMID does not describe that study; (3) the Ashton Manual benzodiazepine-taper percentage is widely used in practice but has no linked primary source in this article and should be sourced or removed; (4) the Hopp et al. abuse-potential abstract citation had a duplicated, incorrect PMID and needs a corrected reference. This article has not received qualified medical review and should not be published until that review, and the source corrections above, are complete.