Quick Answer
A 2026 systematic review and meta-analysis of 13 RCTs involving 1,429 cancer patients found that melatonin supplementation did not produce a statistically significant overall improvement in sleep quality (SMD -0.85, 95% CI -2.46 to 0.76). Heterogeneity was extreme (I² = 94.5%). GRADE evidence was rated very low certainty. Modest benefits were seen in US-based subgroups and at doses up to 6mg. The authors concluded melatonin shows "promising but uncertain benefits" for cancer-related sleep disruption. Verdict: Uncertain.
At a Glance
What this study looked at
Cancer patients experience some of the most severe sleep disruption of any clinical population. Tumour biology, chemotherapy, radiation, psychological distress, pain, and hospitalisation all fragment sleep - often simultaneously. Given this, cancer patients represent an ideal test population for any sleep intervention: if a compound cannot move the needle here, the case for it in healthier populations is harder to make.
Abbasi and colleagues (2026) conducted a systematic review and meta-analysis of randomised controlled trials testing melatonin supplementation specifically for sleep disorders in cancer patients. The study was registered on PROSPERO (CRD42024573948) and published in Current Reviews in Clinical and Experimental Pharmacology (Bentham Science Publishers). It included 13 RCTs screened from 3,998 records, covering 1,429 participants across multiple cancer types: breast (4 trials), NSCLC (2), prostate (1), colorectal (1), advanced or metastatic solid tumours (3), and any cancer (2).
Seven of the 13 studies measured sleep using the PSQI (Pittsburgh Sleep Quality Index) and were included in the primary meta-analysis. The remaining six used other outcome measures and were reviewed narratively.
Study Selection
The key numbers
1. Primary outcome: not statistically significant overall
Across the 7 studies using PSQI, the pooled standardised mean difference (SMD) was -0.85 in favour of melatonin. This sounds meaningful - an SMD of 0.85 would be considered a large effect in most contexts. But the 95% confidence interval ran from -2.46 to 0.76, crossing zero. A confidence interval that crosses zero means the effect could plausibly be zero or even slightly negative. The result was not statistically significant.
The extreme heterogeneity (I² = 94.5%) is the core problem. When I² exceeds 75%, the studies are so different from each other that mathematically pooling them may produce a number that does not represent any of them accurately. The pooled SMD of -0.85 is not a stable estimate of what melatonin does - it is an average of wildly divergent results from studies with different cancers, different doses, different durations, and different populations.
| Subgroup analysis | SMD | 95% CI | I² | Significant? |
|---|---|---|---|---|
| Overall (7 PSQI studies) | -0.85 | -2.46 to 0.76 | 94.5% | No |
| USA-based trials | -0.57 | -0.99 to -0.14 | 31.5% | Yes |
| Duration up to 4 weeks | Not reported | - | - | Trend (P=0.058) |
| Dose up to ~6mg | Not reported | - | - | P=0.049 (dose-response) |
2. The USA subgroup finding
When analyses were restricted to trials conducted in the USA, the result did become statistically significant: SMD -0.57 (95% CI: -0.99 to -0.14), with I² dropping dramatically to 31.5% - a tolerable heterogeneity level. This is the most internally consistent finding in the paper and suggests melatonin may have a modest, real effect in some contexts.
However, the USA subgroup finding should be interpreted cautiously. Subgroup analyses were pre-specified (this is a strength) but the group was small. Differences between US and non-US trials may reflect differences in patient populations, baseline melatonin levels, study conduct quality, or statistical chance. A single subgroup finding does not overturn a non-significant primary outcome.
3. Dose-response: a ceiling around 6mg
A dose-response analysis found a non-linear relationship between melatonin dose and sleep quality improvement. Benefits were observed at doses up to approximately 6mg. Above 6mg, the benefit attenuated (non-linearity test P = 0.049). The studies used doses ranging from 2mg to 20mg.
This is consistent with what pharmacology would predict. Melatonin receptors (MT1 and MT2) can be saturated at relatively low doses. Above receptor saturation, additional dose does not increase signalling - and at high doses, receptor downregulation may blunt the signal further. The 2mg to 5mg range appears to cover most of the available receptor capacity.
"The overall evidence had very low certainty, and while melatonin appears safe and may offer promising but uncertain benefits for sleep in cancer patients, further high-quality studies are needed before definitive conclusions can be made." - Abbasi et al., 2026
4. Duration: shorter trials showed stronger signals
The meta-regression on trial duration approached significance (P = 0.058): trials lasting up to 4 weeks tended to show stronger effects than trials running longer. One interpretation is that melatonin produces acute sleep benefits that diminish over time, possibly through receptor adaptation. Another interpretation is that longer trials allowed more confounders (changes in cancer treatment, disease progression, seasonal variation) to accumulate, diluting the signal.
5. Safety: acceptable, no serious events
Across 13 trials, no serious melatonin-attributable adverse events were reported. The most commonly noted side effects were headache, fatigue, and dizziness - all mild and transient. The safety profile at doses tested (2-20mg) appears acceptable. This does not imply long-term safety has been established; the longest trial was 52 weeks with a very small sample.
Why the primary result was not significant: the heterogeneity problem
An I² of 94.5% is not a minor statistical footnote. It means that 94.5% of the variance in results across studies cannot be explained by chance - it reflects genuine clinical and methodological differences between the trials. In practical terms, some studies found melatonin worked well; others found no benefit or even a trend toward harm. These divergent results cancel each other out in the pooled average.
The sources of heterogeneity are easy to identify: breast cancer patients have different hormonal profiles than NSCLC patients; a 1-week trial captures different biology than a 52-week trial; 2mg of melatonin works differently than 20mg; oral administration differs from extended-release formulations; patients on active chemotherapy have different baseline melatonin suppression than those in remission.
The GRADE rating of "very low certainty" - the lowest tier on the four-point scale - reflects all of this. It means the true effect of melatonin on sleep in cancer could be substantially different from the estimate provided, in either direction.
Verdict: Uncertain
The primary outcome was not statistically significant. Extreme heterogeneity (I² 94.5%) makes the pooled estimate unreliable. GRADE certainty is very low. A modest, significant effect appeared in the USA subgroup and at doses up to 6mg, suggesting there may be real but context-specific benefits. Melatonin appears safe at tested doses. This evidence base does not support melatonin as a reliable, generalised sleep intervention - even for cancer patients, who represent its strongest clinical case. Rated Uncertain rather than Ineffective because the literature is heterogeneous, not uniformly negative; further well-conducted trials may clarify the picture.
Study limitations
Limitations (editorial analysis)
What This Means Tonight
The practical takeaway
Cancer patients represent the population where melatonin's clinical case is at its strongest: severe sleep disruption, biological plausibility via treatment-induced melatonin suppression, and acute clinical need. If melatonin cannot produce a statistically significant, reliable improvement in this population across 13 trials and 1,429 patients, the case for it as a dependable sleep solution in generally healthy adults with non-restorative sleep is weaker still.
This does not mean melatonin is useless. The USA subgroup data, the dose-response curve, and the safety profile all suggest it may have a role in specific, well-defined contexts - particularly at doses near 2-5mg for short durations in populations with genuinely disrupted circadian timing (jet lag, shift work, acute circadian misalignment). But "melatonin for sleep" as a broad category claim is not supported by this evidence.
The mechanisms most relevant to non-restorative sleep in healthy adults - elevated evening cortisol, blunted slow-wave sleep architecture, nervous system threat-response activation - are not primarily melatonin problems. Addressing them requires targeting different pathways: stress modulation, GABA support, adenosine clearance, and the parasympathetic shift that makes deep sleep architecturally accessible. This is what Reincarn Night Reboot™'s formula targets.
Built for different mechanisms
Melatonin-free by design. Powered by patented Maizinol UP165.
REINCARN Reincarn Night Reboot™ is formulated with patented Maizinol UP165 (250mg, dose-matched to published research), clinically studied for deep sleep architecture - a different mechanism entirely from the melatonin timing pathway. For people whose sleep problem is depth and recovery, not just onset, this is the relevant distinction.
Order Reincarn Night Reboot™Publisher's Note
This issue was selected because the melatonin category is where most of REINCARN's category-level competition lives. We want readers to have access to the best current evidence on melatonin's efficacy, including when that evidence is unflattering to a widely-used supplement. The primary finding here - no statistically significant overall benefit even in cancer patients - is not a verdict against melatonin in all contexts. It is a reason to hold the evidence to account rather than defaulting to category assumptions. REINCARN Reincarn Night Reboot™ does not contain melatonin; this is a deliberate formulation decision based on the mechanisms we believe are relevant to our customer's sleep problem. This Digest issue is published by Zandra Life Sciences Pvt. Ltd., the maker of REINCARN Reincarn Night Reboot™. We have a commercial interest in the melatonin-free positioning, and we disclose it.
Every issue, one paper. No noise.
The Sleep Performance Digest covers one peer-reviewed paper per issue - chosen for study design, journal quality, and relevance to sleep recovery. Published when the science warrants it.
Frequently Asked Questions
Does melatonin improve sleep quality in cancer patients?
Based on a 2026 systematic review of 13 RCTs (N=1,429), melatonin did not produce a statistically significant overall improvement in sleep quality (SMD -0.85, 95% CI -2.46 to 0.76). GRADE evidence was rated very low certainty. A modest significant effect was observed in US-based subgroups and at doses up to 6mg. Overall verdict: Uncertain.
What was the primary finding of the Abbasi et al. 2026 melatonin meta-analysis?
The primary outcome was PSQI score, meta-analysed across 7 of the 13 included studies. The overall SMD was -0.85 favouring melatonin, but the 95% confidence interval (-2.46 to 0.76) crossed zero, meaning the result was not statistically significant. Heterogeneity was extreme at I² = 94.5%, making the pooled estimate unreliable.
What dose of melatonin is most effective for sleep?
In the Abbasi 2026 dose-response analysis, sleep quality improvements were observed with melatonin up to approximately 6mg. Benefits attenuated above 6mg (non-linearity test P=0.049). This is consistent with receptor saturation pharmacology: MT1 and MT2 receptors are likely saturated at low to mid doses, and additional dose does not proportionally increase signalling.
Why was there such high heterogeneity across melatonin sleep studies?
The I² of 94.5% reflects genuine clinical differences between trials: different cancer types (breast, prostate, colorectal, NSCLC, metastatic), doses ranging from 2mg to 20mg, durations from 1 to 52 weeks, different concurrent treatments, and different patient baseline profiles. When studies are this clinically diverse, pooling their results produces a number that may not accurately represent any individual context.
Is melatonin safe for cancer patients?
Across 13 RCTs in this review, no serious melatonin-attributable adverse events were reported. Common side effects were headache, fatigue, and dizziness - all mild. The safety profile appears acceptable at doses studied (2-20mg). Long-term safety data beyond 52 weeks is limited. Cancer patients should consult their oncologist before starting melatonin or any supplement.
If melatonin doesn't reliably improve sleep in cancer patients, what does?
CBT-I (Cognitive Behavioural Therapy for Insomnia) has the strongest evidence base for cancer-related insomnia. Exercise has consistent supporting evidence. A separate 2026 meta-analysis found mind-body practices (yoga, mindfulness, Pilates) produced significant PSQI improvements in perinatal women. For non-cancer adults with non-restorative sleep, the most relevant mechanisms appear to involve stress modulation and nervous system downregulation rather than the melatonin timing pathway.
Source
- Abbasi H, et al. "Safety and Efficacy of Melatonin for Sleep Disorders in Cancer: A Systematic Review and Meta-Analysis of Randomized Clinical Trials." Current Reviews in Clinical and Experimental Pharmacology. 2026; Early View. doi: 10.2174/0127724328470308260825053321. PROSPERO: CRD42024573948. Received: 04 Feb 2026; Accepted: 29 Jun 2026.