Cold Sores (Herpes Labialis): Independent Evidence on Antivirals

by
Key takeaways
  • Cold sores (herpes labialis) are caused by herpes simplex virus type 1 (HSV-1). The WHO estimates about 3.8 billion people under 50, or 64%, carried HSV-1 in 2020, and says herpes is "treatable but not curable" (WHO fact sheet, May 2025).
  • The most recent Cochrane treatment review (search to September 2025) found that aciclovir cream, penciclovir cream and valaciclovir tablets probably shorten healing by about one day compared with placebo. Oral aciclovir tablets may make little or no difference (Lin et al., Cochrane 2026).
  • The reviewers say the clinical importance of that one-day gain is uncertain, because no minimally important difference has been defined for healing time. Unwanted events were uncommon across treatments (Cochrane 2026).
  • Lysine has no reliable preventive effect. The 2015 Cochrane review on preventing cold sores found no evidence that it prevents them (Chi et al., Cochrane 2015).
  • Lemon balm has the most promising natural-remedy signal, but it rests on a handful of small trials. The main 1999 trial of a lemon balm cream (Lo-701) has no sponsor we could confirm, and its reported sample size differs between sources (Koytchev et al. 1999; Khozeimeh et al. 2021 review).
  • Money trail: Aciclovir and valaciclovir are made by GlaxoSmithKline (GSK), which the EU regulator names as the valaciclovir marketing holder (EMA referral report, 2010). Of the 87 trials in the 2026 Cochrane review, 43 were funded by industry. The two US trials of penciclovir (1997) and valaciclovir (2003) that we reviewed do not name a sponsor in the records we could open, so their funding remains unverified (Spruance et al. 1997; Spruance et al. 2003).
  • Evidence grade: Moderate that aciclovir and valaciclovir shorten healing a little. Weak for lemon balm. Insufficient for lysine and for lasers, zinc patches and propolis. Risk to the eyes and to newborns if the virus is spread carelessly.

Independent evidence review · Prescription medicine

Antiviral creams and valaciclovir tablets probably shorten a cold sore by about one day when started at the first tingle, and the independent review found few side effects. The effect is small, and most of the trials behind it were funded by drug makers. Lysine has no reliable preventive effect, and lemon balm rests on a few small trials. Our confidence: moderate that aciclovir, penciclovir and valaciclovir help a little; low for lemon balm; very low for lysine. The virus itself cannot be cleared, so these treatments shorten an episode rather than cure the infection.

Best evidence for about one day shorter healing with aciclovir or penciclovir cream, or valaciclovir tablets, when started early
Main risks rare side effects in trials; spread by skin contact; eye infection if cream reaches the eye; neonatal herpes if a baby is kissed during an active sore
Key rule the evidence points to starting treatment at the first tingling, before blisters form; see a clinician for eye involvement, pregnancy, a weakened immune system or frequent recurrences
Safety first
See a clinician promptly if a sore is in or near your eye, if the eye becomes painful or red, or if you have a weakened immune system, eczema covering a large area of skin, a high fever, or a sore that will not heal. Herpes near the eye can damage sight. Do not kiss a baby while you have an active cold sore, because herpes can cause neonatal herpes, a severe infection in newborns (NHS). Keep antiviral cream away from the eyes, and check the product leaflet for its own warnings. Nothing in this article replaces a clinical assessment.

Table of contents

Evidence summary

Claim Evidence Source Funding / conflict Strength
Aciclovir cream shortens healing Probably about one day sooner than placebo. Seven studies with 3,475 people for the healing outcome, and no increase in unwanted events (Cochrane 2026). Lin et al., Cochrane 2026 Review has no dedicated funding. Most of the underlying trials were industry-funded (43 of 87 studies across the review). Moderate
Penciclovir cream shortens healing Probably about one day sooner. The healing data rest on one study of 528 people (Cochrane 2026). The larger 1997 US trial enrolled 2,209 people, and reported a median of 4.8 days to heal with penciclovir against 5.5 days with vehicle cream. Cochrane 2026; Spruance 1997 The 1997 trial's record names no sponsor, and the trial was run under the name "Topical Penciclovir Collaborative Study Group". Sponsor not verified. Moderate
Valaciclovir tablets shorten healing Probably about one day sooner. Two studies with 1,218 people for the healing outcome (Cochrane 2026). The 2003 trials used 2 g twice daily for one day, started at the first symptoms. Cochrane 2026; Spruance 2003 The 2003 record names no sponsor or funding. The product's marketing holder is GSK (EMA 2010). Sponsor of the trials not verified. Moderate
Oral aciclovir tablets shorten healing May make little or no difference. One study with 145 people (Cochrane 2026). Cochrane 2026 Not reported in the summary we could open. Weak (probably no benefit)
Long-term antivirals reduce how often cold sores come back Oral aciclovir 400 mg twice daily reduced recurrences (risk ratio 0.26, 95% CI 0.13 to 0.51; 177 people). Daily valaciclovir 500 mg gave a small reduction. Short courses did not prevent recurrences (Cochrane 2015). Chi et al., Cochrane 2015; JMIR Dermatology summary, 2022 Of 32 trials in the 2015 review, 18 were manufacturer-funded, 4 non-profit, 10 unknown. Moderate (for frequent recurrences)
Lysine prevents cold sores No evidence of prevention in the Cochrane review. Long-term 1,000 mg daily was no different from placebo in the one summarised trial, and the evidence was rated very low quality. Chi et al., Cochrane 2015; JMIR summary Lysine is an unpatented amino acid sold by many companies, so no single trial sponsor dominates. Trial funders were mostly not reported. Insufficient
Lemon balm cream shortens symptoms The 1999 Lo-701 trial reported symptom and lesion benefit against placebo. A 2021 systematic review of three trials found lemon balm reduced pain and lesion size more than acyclovir or placebo, with no side effects reported. Koytchev et al. 1999; Khozeimeh et al. 2021 The 1999 sponsor could not be confirmed, and no funding is listed in the indexed record. The 2021 review has no funding statement. Weak
Propolis, zinc patches, photodynamic therapy, lasers Some signal of benefit in the Cochrane review, but the evidence is limited and the trials are small. Cochrane 2026 Not reported in the summary we could open. Insufficient
Antivirals cause serious side effects Unwanted events were uncommon. Valaciclovir had no serious unwanted events in the two trials, and aciclovir cream showed no increase versus placebo (Cochrane 2026). Cochrane 2026 See funding of the included trials above. Moderate

Independent evidence and credibility scorecard

Tier 1 is independent of the subject, with no identifiable financial stake in the outcome. Tier 2 has indirect or unknown ties. Tier 3 is an interested party, such as a product trial with an unidentified sponsor. Tier 4 is produced or funded by the maker or seller. Credibility A is high, B is solid but can be tilted by incentives, C is interested, and D is self-interested.

Source Who funds it Country Independence tier (1–4) Credibility (A–D) Why it would tell the truth / residual bias
Lin et al., Cochrane Library 2026 (CD015216) No dedicated review funding. Included trials: 43 industry, 25 non-profit, 1 none, 18 not reported. International; review authors' countries not shown on the summary page 1 A Its usefulness depends on being accurate and trusted by clinicians. Residual bias: most of the trials it pools were drug-funded, and many are small.
Chi et al., Cochrane 2015 (CD010095) Review funding not shown on the summary page. Included trials: 18 manufacturer, 4 non-profit, 10 unknown. International; not shown 1 B Same accuracy incentive as the 2026 review, but older. Residual bias: the prevention trials are older and some outcomes were rated low quality.
WHO herpes simplex fact sheet (2025) Intergovernmental UN agency. Funding sources not checked in detail. Switzerland (Geneva headquarters) 1 B Public-health mandate, with no product to sell. Residual bias: its infection figures are modelled estimates, and its treatment statements are brief.
NHS: cold sores (England) Publicly funded health service UK 1 B Legal duty to give public guidance. Residual bias: patient-facing guidance is simplified and does not name every antiviral.
European Medicines Agency: Valtrex referral assessment (2010) EU regulator. The marketing holder named in the report is GSK. The agency's own funding model was not checked for this article. EU 2 B Legal accountability for medicine safety and labelling. Residual bias: the report deals with label harmonisation for a GSK product, not with trial efficacy, and the holder details date from 2010.
Khozeimeh et al., systematic review of lemon balm vs acyclovir (2021) No funding or conflict statement found. Three trials from 955 articles. Iran 2 C Academic. Journals reward positive findings. Residual bias: three small trials, no effect sizes, and no funding disclosure.
Sivesind et al., JMIR Dermatology summary (2022) No funding statement in the text we could open. One co-author holds a fellowship from Pfizer, disclosed in the article. Not verified 2 C Academic summary with a disclosed industry tie. Residual bias: the disclosed industry tie and editorial roles could shape which options the summary emphasises.
Koytchev et al., Lo-701 lemon balm trial (1999) Sponsor could not be verified. No funding listed in the indexed record. The cream is a branded product. Not verified 3 C A product trial with an unidentified sponsor. Residual bias: a positive result helps a branded herbal product, and the sample size is reported differently across sources.
Spruance et al., penciclovir cream trial (JAMA 1997) Sponsor not verified. The record names no sponsor, and the trial ran under a "Collaborative Study Group" name. United States (31 ambulatory sites) 4 D A product trial of a topical drug, provisionally treated as maker-linked until the sponsor is confirmed. Residual bias: healing time depends on how lesions are scored, and the trial's sponsor affects how the result is framed.
Spruance et al., valaciclovir trials (Antimicrob Agents Chemother 2003) Sponsor not verified. The record has no funding entry. The product's holder is GSK (EMA 2010). Not verified 4 D A product trial of the maker's drug, provisionally treated as maker-linked. Residual bias: a short high-dose course is a commercially useful result, so the trial's framing deserves scrutiny (an inference, not a documented fact).

Convergence and clustering. The strongest signal is the agreement between the independent Cochrane review and the drug-funded trials on the same direction of effect: about one day shorter healing for the three antivirals. The independent review is also cautious about size, which the drug-funded trials were not. The antiviral trials we reviewed were run in the US, and the country of the 1999 lemon balm trial was not verified, so the picture rests heavily on one country's trial base.

What a cold sore is, and the treatments available

A cold sore, also called herpes labialis or a fever blister, is a blister-like sore on or around the lips. Most are caused by herpes simplex virus type 1 (HSV-1). The WHO says the virus spreads mainly through contact with sores, saliva or skin around the mouth, and that people can pass it on from skin that looks normal, though the risk is highest when sores are active (WHO). The virus stays in the body after the sore heals. Sores come back in people who carry it, often when they are unwell, in strong sunshine, or during a period (NHS).

The NHS says a typical cold sore clears within about 10 days, and that a person is contagious from the first tingling until the sore has fully healed (NHS). The NHS says a pharmacist can recommend an antiviral cream (NHS). Treatments fall into four groups:

  • Topical nucleoside antivirals: aciclovir cream and penciclovir cream, applied to the lip several times a day.
  • Oral antivirals: aciclovir and valaciclovir tablets, usually prescribed for severe or very frequent episodes. Valaciclovir is the prodrug of aciclovir, and the EU referral report names GSK as its marketing holder (EMA, 2010).
  • Natural and herbal options: lemon balm (Melissa officinalis) cream and lysine supplements, which many people try.
  • Other options: photodynamic therapy, laser treatment, zinc patches and topical propolis. The Cochrane review found these may add some benefit, but the evidence is limited (Cochrane 2026).

Antiviral medicines do not clear the infection. The WHO describes herpes as treatable but not curable, and says medicines can shorten outbreaks and reduce their severity (WHO).

How the treatments work

Aciclovir, penciclovir and valaciclovir are antiviral nucleoside analogues that interfere with the herpes virus copying its DNA inside infected cells. We did not open a pharmacology source for this article, so treat that description as general background. The practical consequence is well supported, though: the trials that show benefit treated the episode in its first hours or days, while the virus was still copying in the lip (Cochrane 2026; Spruance et al. 2003).

Lemon balm's proposed mechanism is laboratory-based. We did not open the laboratory papers for this article, and we found no human study that measures viral activity in lemon balm users, so the mechanism is unproven on the lip. Lysine is usually said to compete with arginine, a building block the virus needs. That idea was not checked against trial data here, and the Cochrane review found no preventive effect (Chi et al., Cochrane 2015).

What works and what does not

Treatment What the evidence shows Status
Aciclovir cream, applied early About one day shorter healing against placebo (Cochrane 2026) Works, modestly
Penciclovir cream, applied early About one day shorter healing against placebo (Cochrane 2026); one study for the healing outcome Works, modestly
Valaciclovir tablets for episodes About one day shorter healing against placebo; two studies (Cochrane 2026) Works, modestly
Long-term aciclovir or valaciclovir to prevent recurrences Reduces how often sores come back, by a modest amount (Cochrane 2015) Mixed (for frequent recurrences)
Oral aciclovir tablets for episodes May make little or no difference (one study, 145 people) Probably no benefit
Lemon balm cream Small trials show benefit against placebo or acyclovir; the sponsor and sample size are unclear Promising, unproven
Lysine supplements No reliable preventive effect in the Cochrane review Does not work as shown
Propolis, zinc patches, laser, photodynamic therapy Limited, small studies Insufficient

Benefits by claim

Does aciclovir or penciclovir cream make a cold sore heal faster?

Probably, by about one day. The 2026 Cochrane review pooled seven studies with 3,475 people for the healing outcome for aciclovir cream, and found it probably shortens healing by about a day with no increase in unwanted events. For penciclovir cream, the review found one study of 528 people for healing (Cochrane 2026). The review also found that the two creams may perform similarly. Its authors caution that the clinical value of the one-day reduction is uncertain, because no minimally important difference has been set for healing time.

The older US penciclovir trial reported the same direction of effect. Its record (2,209 people enrolled, 1,573 who started treatment for a recurrence) shows the effect on healing, but the full abstract and sponsor were not opened (Spruance et al. 1997). The NHS advises starting creams at the first tingle, because they may not work once blisters have formed (NHS).

Do valaciclovir tablets work faster than cream?

The valaciclovir trials suggest they do, when taken early. The 2026 Cochrane review found two studies with 1,218 people for the healing outcome, and found valaciclovir probably shortens healing by about one day (Cochrane 2026). The 2003 trials used 2 g twice daily for one day, started at the first symptoms. Each study enrolled roughly 1,500 to 1,600 people, and in the first study the episode duration was about one day shorter with the one-day regimen (Spruance et al. 2003). The trials were not designed to show that the lip heals completely faster, and the clinical gain is modest.

Oral aciclovir tablets are different. The one study in the review (145 people) found little or no effect on healing time, so the review does not support them for episodes in people who are otherwise well (Cochrane 2026).

Can antivirals stop cold sores from coming back?

For people who get frequent episodes, long-term oral treatment reduces recurrences. The 2015 Cochrane review pooled 32 trials and found that oral aciclovir 400 mg twice daily reduced recurrences, with a risk ratio of 0.26 (95% CI 0.13 to 0.51; 177 people). Daily valaciclovir 500 mg gave a small reduction. Short courses did not prevent recurrences (Chi et al. 2015; JMIR Dermatology summary). The 2015 review's funding is also worth noting: 18 of its 32 trials were funded by the drug manufacturer, and the funding for 10 was unknown. Long-term treatment is a clinical decision for a prescriber, and it carries a continuing cost.

Does lysine stop cold sores?

No reliable evidence shows that it does. The 2015 Cochrane review found no evidence that lysine prevents cold sores. The JMIR summary of that review reports that long-term oral lysine at 1,000 mg daily showed no significant difference from placebo, and rated the evidence as very low quality (Chi et al. 2015; JMIR summary). The trials are older and small. Some individuals may feel their sores improve, but a placebo-controlled trial is needed to tell that apart from natural variation. The NHS page on cold sores does not recommend lysine (NHS).

Does lemon balm help?

Possibly, but the evidence is thin. The main positive trial is Koytchev and colleagues' 1999 study of Lo-701, a balm mint extract (Koytchev et al. 1999). It was reported as a double-blind, placebo-controlled trial of a 1% cream of a lyophilised aqueous extract, Lomaherpan. The sources we could find describe the outcome differently. One reports a significant difference in lesion size and blisters by day two. A more cautious account reports lower symptom scores by day two, with the five-day trend not significant. Sample sizes also differ: one summary says 66 patients were analysed, while another says 120 patients who met inclusion criteria were randomised. We could not open the full paper to resolve this.

A 2021 systematic review by Khozeimeh and colleagues screened 955 articles and kept three trials with adequate reporting. It found lemon balm more effective than acyclovir or placebo at reducing lesion pain and size, with no significant effect on inflammation and no side effects. It gives no effect sizes or confidence intervals (Khozeimeh et al. 2021). Three small trials are not enough to recommend lemon balm over tested antivirals. The Cochrane review did not include lemon balm as a treatment with a clear benefit.

Are propolis, zinc, lasers or light therapy useful?

The Cochrane review says photodynamic therapy, laser treatment, zinc patches and topical propolis may offer some benefit, but the evidence is limited, and it gives no numbers in the summary we could open (Cochrane 2026). The NHS notes that light or laser devices have little evidence behind them (NHS).

Risks and all side effects

Eye infection
Herpes in or near the eye can cause serious damage. Keep cream away from the eyes, and see an eye specialist or your GP the same day for eye pain, redness, light sensitivity or blurred vision. This is general clinical advice; the sources we opened for this article do not give eye-specific guidance.
Newborns
Herpes can cause severe infection in a newborn. Avoid kissing babies while a cold sore is active, and wash your hands after touching a sore (NHS).

In the Cochrane review, unwanted events were uncommon across all treatments. Aciclovir cream showed no increase in serious or mild-to-moderate unwanted events against placebo, and valaciclovir had no serious unwanted events in the two trials, with mild events similar to placebo (Cochrane 2026). Rarer reactions to oral antivirals are listed in product labels. We did not open a current label for this article, so ask a pharmacist. The trials were short, so rare or long-term effects may not show up in them.

Herbal and supplement products carry their own risks. The sources we opened do not give safety data for long-term, high-dose lysine. The 2021 lemon balm review reports no side effects, but it gives no numbers.

All interactions

We did not find a complete interaction list in the sources we opened, and we did not open a current drug label for each product. The following points are general and should be checked with a pharmacist:

  • Oral aciclovir and valaciclovir: interaction and kidney-dose questions belong with a pharmacist, who can check the current product label.
  • Lemon balm: the trials we reviewed did not test it alongside other medicines.
  • Lysine: one secondary summary of lysine trials advises caution in people with heart or gallbladder disease. We did not verify that summary against the original papers.

Antivirals and cold sore products are not a substitute for prescribed medicines. Tell your pharmacist about all products you use.

Who should be seen by a clinician first

  • People with a weakened immune system. The NHS says a GP may prescribe antiviral tablets if sores are very large, painful or keep coming back (NHS). Immune-suppressed people need that assessment rather than self-treatment.
  • Pregnant and breastfeeding people. We did not check pregnancy guidance for these products in this article. Ask a clinician.
  • Babies and young children. Do not kiss a baby during an active sore (NHS), and seek advice for any baby with sores or a fever.
  • People with eczema, where the risk of herpes spreading over broken skin is a reason for prompt assessment. Our separate eczema guide is linked in related research.
  • People with kidney disease, for whom oral antiviral doses should be checked with a clinician.

Doses studied and how the products are used

These are doses studied in trials, not personal instructions.

  • Aciclovir cream (5%): studied in the Cochrane trials as a topical treatment applied several times a day for the early phase of an episode.
  • Penciclovir cream (1%): the 1997 US trial applied the cream every two hours while awake for four days, starting within an hour of symptoms (Spruance et al. 1997).
  • Valaciclovir tablets: the 2003 trials studied 2 g twice daily for one day, started at the first symptoms, and a two-day version (2 g twice daily for one day, then 1 g twice daily for one day) (Spruance et al. 2003).
  • Long-term prevention: aciclovir 400 mg twice daily and valaciclovir 500 mg once daily were the doses studied in the 2015 review (Chi et al. 2015).
  • Lemon balm cream (Lo-701): a 1% cream of a dried aqueous extract applied for five days in the 1999 trial. The application frequency is reported differently across sources.
  • Lysine: 1,000 mg daily was the long-term dose studied in the 2015 review. Higher doses have been studied in small trials, which we did not verify here.

Practical points that the NHS gives are simple: start antiviral cream at the first tingle, not after blisters form; wash hands after touching the sore; keep cream away from the eyes; and avoid kissing or oral sex until the sore has fully healed (NHS). Pain relief such as paracetamol or ibuprofen can also help. Aspirin is not advised for under-16s (NHS).

Product quality. Cream and tablet antivirals are regulated medicines with batch control. Lemon balm products sold as creams or lip balms vary in their extract content, and we found no independent testing of Lo-701 concentrations in the sources we opened.

Follow the money: who makes it and who funded the evidence

Who makes the medicines

Aciclovir (Zovirax). The brand is now associated with GlaxoSmithKline, which manufactures aciclovir under the Zovirax name, according to a reference summary we opened (EBSCO). The drug was originally patented by Burroughs Wellcome in 1974, and Wellcome's own records describe its development in the 1980s (Wikipedia summary). The current holder for each country varies, and we did not check every national register.

Valaciclovir (Valtrex). The EMA's 2010 referral report lists GlaxoSmithKline and its associated companies as the marketing authorisation holder for Valtrex in the EU (EMA 2010). Because that report is from 2010, the holder details should be checked on the current EMA product page. GSK's headquarters country was not verified in the sources we opened.

Penciclovir (Denavir). One source we found names SmithKline Beecham as the maker of penciclovir cream under the Denavir name. We did not confirm that from the trial record, which names no sponsor, so the current holder is unverified.

Lemon balm cream (Lomaherpan). Pharmacy listings name Infectopharm Arzneimittel as the manufacturer of Lomaherpan lip cream, made with a concentrated lemon balm extract (Shop Apotheke listing). An Austrian product leaflet names InfectoPharm Arzneimittel und Consilium GmbH as the marketing authorisation holder (pharmacy listing), and the company is based in Heppenheim, Germany. The company describes itself as family-owned, founded in 1988, and now in its second generation (Apotheke Adhoc; InfectoPharm fact sheet, 2026). A US trademark record shows a LOMAHERPAN mark filed in 1984 and held by Lompharm Rudolf Lohmann GmbH KG (trademark record). We could not confirm how the mark passed to the current company, so that link is historical only. We could not confirm that the 1999 Lo-701 trial was sponsored by the current company.

Lysine. Lysine is an unpatented amino acid sold by many supplement companies. We did not find a single owner or a dominant brand in the sources we opened.

Who funded the evidence

The clearest funding data come from the Cochrane reviews. Of the 87 studies in the 2026 review, 43 were funded by the pharmaceutical industry, 25 by non-profit organisations, one had no funding, and 18 did not report funding (Cochrane 2026). The review itself had no dedicated funding. The 2015 review found 18 of 32 trials were manufacturer-funded (Chi et al. 2015).

The individual trials we reviewed are harder to trace. The 1997 penciclovir trial and the 2003 valaciclovir trials have no sponsor or funding entry in the indexed records we could open, and the full texts were not accessible to us. The 1999 Lo-701 trial also has no funding entry, and its sponsor could not be confirmed. The 2021 lemon balm review has no funding statement. Where a trial's sponsor is unknown, we treat it as provisionally interested, and we have said so in the scorecard. We did not find a regulatory warning, recall or lawsuit for these products in the sources we reviewed, but we did not run a full enforcement search across every regulator.

Two points cut the other way. The WHO and NHS sources are public and have no product to sell, and the Cochrane 2026 review's direction matches the drug-funded trials. Those facts strengthen the case that the direction of the effect is real, while the size of the effect remains uncertain.

Follow the money

Research funding at a glance

Funding & backersSource & studyClaim & limits

10 sources graded. The counts below are the independence tiers assigned in the scorecard above. They count sources, not funding amounts or evidence quality.

Tier 14Independent
Tier 23Indirect ties
Tier 31Interested party
Tier 42Self-interested

Public funding alone does not prove independence, and commercial funding alone does not prove a claim false. Named funders, countries and relationships are in the scorecard and source notes.

Frequently asked questions

Do cold sores go away on their own?

Yes. The NHS says a cold sore usually clears within about 10 days, and that the person is contagious until it has fully healed (NHS). Antivirals shorten the episode by about a day, according to the 2026 Cochrane review (Cochrane 2026).

Is aciclovir cream worth using?

The evidence says it probably shortens healing by about a day, with no increase in unwanted events. The effect is modest, and it works best started at the first tingle (Cochrane 2026; NHS).

Is cream better than tablets?

The Cochrane review does not show that cream beats tablets. Valaciclovir tablets probably shorten healing by about one day, and so do aciclovir and penciclovir creams. Oral aciclovir tablets may make little or no difference (Cochrane 2026). Tablets are usually reserved for severe or frequent episodes (NHS).

Does lysine work for cold sores?

Not on the evidence we found. The 2015 Cochrane review found no evidence that lysine prevents cold sores, and the trials are small and older (Chi et al. 2015).

Does lemon balm cure cold sores?

No treatment cures the virus. Lemon balm has shown some benefit in small trials, including a 1999 trial of a lemon balm cream (Koytchev et al. 1999), but the sponsor, sample size and outcome differ across sources, and a 2021 systematic review of three trials gives no effect sizes (Khozeimeh et al. 2021).

Can cold sores spread to the eyes?

Yes. Herpes in or near the eye can cause serious damage. Keep antiviral cream away from the eyes, and seek same-day medical advice for eye pain, redness or blurred vision (Cochrane 2026; NHS).

What triggers a cold sore?

The NHS lists illness, sunshine, sunbeds and periods as possible triggers (NHS). The NHS also suggests an SPF 15+ lip balm outdoors. We did not find a trial that proves sun protection prevents episodes.

Can I prevent cold sores from spreading to other people?

The NHS advises washing hands after touching a sore, not kissing anyone or having oral sex until the sore has fully healed, and not kissing babies during an active sore (NHS). The WHO notes that people can pass the virus on from skin that looks normal, though the risk is highest when sores are active (WHO).

Who pays for the research on these treatments?

For the antivirals, most trials were drug-funded. The 2026 Cochrane review counted 43 industry-funded studies out of 87, and 18 with no reported funding (Cochrane 2026). Funding does not prove that a result is wrong, and the independent review reached the same direction of effect.

Sources and funding notes

What we could not verify. The full text of the 1997 penciclovir trial (JAMA) and the 2003 valaciclovir trials was not accessible to us, so the sponsor of each is unverified. The PubMed page for the 1997 trial was rate-limited, and we used its indexed record, with the healing figures taken from the trial abstract as summarised in search results. The full text of the 1999 Lo-701 trial was not accessible, so its sample size (66 analysed or 120 randomised, according to different sources) and its funder remain unresolved. We could not open the Cochrane review's author declarations, or the funding statement of the 2021 lemon balm review, which has none. We did not open the current NHS or EMA product labels, and we did not check national registers for the current holder of each antiviral. The GSK headquarters country and the current holder of penciclovir (Denavir) are unverified.

  1. Lin HS, Lin PT, Shih IC, Lu ML, Tsai YS, Chi CC. Interventions for treatment of herpes labialis (cold sores on the lips). Cochrane Database of Systematic Reviews 2026, Issue 9, CD015216. Systematic review; no dedicated review funding; included trials 43 industry, 25 non-profit, 1 none, 18 not reported. Cochrane summary
  2. Chi CC, Wang SH, Delamere FM, Wojnarowska F, Peters MC, Kanjirath PP. Interventions for prevention of herpes simplex labialis (cold sores on the lips). Cochrane Database of Systematic Reviews 2015, Issue 8, CD010095. Systematic review; included trials 18 manufacturer, 4 non-profit, 10 unknown. Cochrane summary
  3. Sivesind T, Viola J, Zhang L, Dellavalle R, Chi C-C. JMIR Dermatology summary of the 2015 Cochrane prevention review, 2022. Academic summary; one co-author holds a Pfizer fellowship (disclosed); no funding statement in the text we opened. PMC
  4. World Health Organization. Herpes simplex virus fact sheet, 30 May 2025. Intergovernmental public-health body; funding not checked in detail. WHO
  5. World Health Organization. Globally, an estimated two-thirds of the population under 50 are infected with herpes simplex virus type 1 (news release, 28 October 2015). Intergovernmental; estimate published in PLOS ONE. WHO news
  6. NHS England. Cold sores (herpes simplex) guidance page. Publicly funded health service. NHS
  7. European Medicines Agency. Valtrex (valaciclovir) referral assessment report, 2010. EU regulator; marketing holder named as GlaxoSmithKline and associated companies at the time. EMA
  8. Spruance SL, Rea TL, Thoming C, Tucker R, Saltzman R, Boon R. Penciclovir cream for the treatment of herpes simplex labialis: a randomized, multicenter, double-blind, placebo-controlled trial. JAMA 1997;277(17):1374-1379 (PMID 9134943). Product trial by a Topical Penciclovir Collaborative Study Group; sponsor not verified. PubMed
  9. Spruance SL, Jones TM, Blatter MM, Vargas-Cortes M, Barber J, Hill J, Goldstein D, Schultz M. High-dose, short-duration, early valacyclovir therapy for episodic treatment of cold sores. Antimicrobial Agents and Chemotherapy 2003;47(3):1072-1080 (PMID 12604544; PMC149313). Two randomised product trials; sponsor not verified in the record. PubMed
  10. Koytchev R, Alken RG, Dundarov S. Balm mint extract (Lo-701) for topical treatment of recurring herpes labialis. Phytomedicine 1999 (PMID 10589440). Product trial of a lemon balm cream; sponsor could not be verified. PubMed
  11. Khozeimeh F, Mojtahedi N, Seyedmoalemi Z. Comparison of Melissa with acyclovir on recurrent labial herpes (a systematic review). Journal of Isfahan Dental School 2021;17(2):216-224. Academic review of three trials; no funding statement found. Journal
  12. EBSCO Research Starters. Aciclovir (ACV) entry. Reference summary; manufacturer named as GSK under the Zovirax brand. EBSCO
  13. InfectoPharm Arzneimittel und Consilium GmbH. Company fact sheet, February 2026. Company's own description; family-owned, Heppenheim. Fact sheet
  14. Apotheke Adhoc. Feature on InfectoPharm's history (trade press). Company history; trade press. Apotheke Adhoc
  15. Shop Apotheke. Lomaherpan lip-care cream listing. Pharmacy retailer; names the manufacturer. Shop Apotheke
  16. Trademark record for LOMAHERPAN (filed 1984), owner Lompharm Rudolf Lohmann GmbH KG. Public trademark register; historical ownership only. Justia

Educational research reviewed 9 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.

Have a question — or want us to cover something?

Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.

We store your topic, message, optional email, and this page so we can manage and reply to the request. Do not include diagnoses, medications, or other sensitive medical information. See our Privacy Policy.