Oral Cancer: The Early Signs You Can See in a Mirror

In short

Any sore, lump, white patch or red patch in the mouth that lasts more than two weeks needs looking at. Here is what the patches mean, why a biopsy beats waiting, and what changes after stopping.

Emergency if: Go to emergency care now for difficulty breathing or noisy breathing, difficulty swallowing your own saliva or drooling because you cannot swallow, bleeding from the mouth that will not stop with pressure, or rapid swelling of the mouth, floor of the mouth or neck. These threaten the airway, and an airway problem is measured in minutes rather than days.

On this page

What should I be looking for in my own mouth?

Anything that has not healed in two weeks. That is the whole rule, and NIDCR states it plainly: if you have any of these symptoms for more than two weeks, see a dentist or a doctor. The list is a sore, irritation, lump or thick patch in the mouth, lip or throat; a white or red patch in the mouth; a persistent sore throat, a feeling that something is caught in the throat, hoarseness or loss of the voice; a lump in the neck; difficulty chewing, swallowing or speaking; difficulty moving the jaw or tongue; swelling of the jaw that makes dentures fit badly; pain or bleeding in the mouth; numbness in the tongue or elsewhere in the mouth; and ear pain.

Looking properly takes about a minute. Use daylight or a bright lamp and a mirror, take out any dentures, and go in the same order every time: the lips inside and out, the gums, the inside of both cheeks pulled outward, the roof of the mouth, the tongue on top, then the tongue lifted to see underneath and pushed side to side to see both edges, then the floor of the mouth. Finish by running your fingers down both sides of the neck and under the jaw, feeling for a lump. The edges of the tongue and the floor of the mouth are the areas most worth the extra light.

Two weeks is chosen because ordinary mouth injuries settle inside it. A cheek bitten while eating, an ulcer from a sharp tooth edge or a denture rubbing, a burn from hot tea: these heal. Anything that outlasts the healing time of ordinary damage is no longer ordinary damage, whatever else it might turn out to be. Most such patches turn out not to be cancer, and that is something only an examination can tell you; waiting longer never tells you. The point of the two-week rule is that it costs one appointment to find out, and the cost of not finding out is measured in stages.

A white patch or a red patch: which one is more dangerous?

The red one, by a wide margin. StatPearls defines leukoplakia as a white patch or plaque in the mouth that cannot be wiped off, and erythroplakia as a firm red patch or plaque; both are usually non-tender, which is exactly why they get ignored. A patch that does not hurt does not feel like a problem. In a meta-analysis cited by StatPearls, the rate of malignant transformation was 9.5% for leukoplakia and 33.1% for erythroplakia, against 7.9% for all oral premalignant disorders combined.

Those figures deserve a moment. Roughly one in three red patches goes on to become cancer, and roughly one in ten white ones. That still means most white patches do not, which is worth saying to anyone who has just found one and is frightened. But it also means neither is something to watch at home. Oral premalignant lesions are estimated to occur in roughly 1.5% to 4.5% of the world's population and are disproportionately common in men, so these are not rare findings.

Where the patch sits matters as much as its colour. StatPearls identifies the lateral border of the tongue and the floor of the mouth as high-risk locations, and these are also the two places hardest to see without lifting and turning the tongue deliberately. A patch on the inside of the lower lip and a patch on the side of the tongue are not the same finding, and the second one earns a faster appointment.

Why does tobacco in every form, betel quid and areca nut matter?

Because together they account for about a third of the disease worldwide, and that share tracks wherever these products are part of ordinary daily life. WHO puts the global burden of cancers of the lip and oral cavity at an estimated 389 846 new cases and 188 438 deaths in 2022, ranking them 13th among cancers worldwide, with tobacco, alcohol and areca nut or betel quid use among the leading causes. IARC's 2024 analysis found that 120 200 of the 389 800 oral cancer cases in 2022 could be attributed to smokeless tobacco and areca nut use, which is 31% of the global total.

This is about products, not about the people who use them, most of whom started young, in company, and long before anyone told them what the packet contained. Smokeless products are not a safer version of smoking, and a quid without tobacco is not a harmless quid, because the nut itself is the Group 1 carcinogen: IARC classifies areca nut as carcinogenic to humans, Group 1, and its main active ingredient arecoline as possibly carcinogenic, Group 2B. Those are facts about chemistry rather than character.

The burden follows the products rather than any particular place or any particular people. IARC estimated that 300 million people use smokeless tobacco and 600 million use areca nut, and that 96.4% of the attributable cases occurred in low- and middle-income countries. Areca nut mattered more than tobacco-only products in that analysis: a higher proportion of oral cancers were due to areca nut products, with or without tobacco. If any of these are part of your day, the next section is the one that matters.

What actually changes after stopping?

The risk falls, and it falls enough to be worth the difficulty of stopping. Reviewing the evidence in IARC Handbooks Volume 19, WHO stated that key measures for significantly reducing the risk of oral cancer include avoiding or stopping tobacco smoking and the use of areca nut, including betel quid with or without tobacco, and reducing alcohol consumption. The same assessment is honest about the limits of the evidence: the causal association between cessation and cancer is established, while the impact of some specific interventions on cancer incidence or death is not fully ascertained.

Two things are worth being straight about. Stopping lowers risk, but the evidence does not support telling a long-term user that risk drops immediately to that of someone who never started. And no mouthwash, oil, paste, vitamin or herbal preparation has been shown to make an established white or red patch go away for good. Anything sold on that promise is being sold, not prescribed. What helps is removing the exposure and having the patch properly examined.

Practical support matters more than willpower. WHO's assessment noted that behavioural interventions using telephone counselling, self-help manuals and tip sheets show a significant impact among adult populations, though the evidence is less conclusive in younger people. A dentist or doctor can also treat the sharp tooth or ill-fitting denture that keeps a patch irritated, which removes a second insult from the same piece of mucosa.

Submucous fibrosis: the stiff mouth that comes before cancer

This is the condition to know about in any area where areca nut is chewed, because it announces itself long before a cancer does. StatPearls describes oral submucous fibrosis as arising from betel nut use, which causes submucosal fibrosis and progressive scarring of the oral cavity, leading to trismus, difficulty swallowing or painful swallowing. Trismus means the mouth will not open as wide as it used to. People often notice it first as trouble taking a large bite, managing a wide yawn, or letting a dentist work at the back of the mouth.

It is a potentially malignant condition rather than a cancer. In the meta-analysis cited by StatPearls, oral submucous fibrosis carried a malignant transformation rate of 5.2%, lower than leukoplakia's 9.5% but far from negligible, and unlike a single patch it affects the whole lining. Estimates of that transformation rate vary between studies, which is a reason to treat any figure as approximate and the condition itself as serious.

The practical consequence is that a mouth which has become stiff is harder to examine and harder to treat later. Someone who cannot open wide cannot be properly inspected, so a lesion at the back can go unseen for a long time. If your mouth opening has narrowed over months or years, that is a reason to be assessed now rather than after a patch appears, and the single most useful action alongside it is stopping the nut.

How is a suspicious patch checked, and why is watching it not enough?

By taking a piece of it and looking at the cells. StatPearls is direct that tissue sampling is warranted to rule out frank malignancy in most cases, and that lesions on the lateral border of the tongue or the floor of the mouth warrant excisional biopsy. A clinical opinion, however experienced, describes what a patch looks like. Only histology says what the cells are doing, and dysplasia, which is the change that matters, is invisible to the eye.

The argument against waiting gets stronger when you look at what biopsies miss. StatPearls cites a study in which removing whole lesions revealed a 7% rate of occult carcinoma that the initial biopsy had not shown, meaning cancer already present but not sampled. If a biopsy can under-call a lesion, then a review appointment in three months, which samples nothing at all, is not a diagnostic plan. It is a delay with a date attached.

Adjunctive tests are often offered instead, and the evidence does not support relying on them. The Cochrane review of oral cancer screening found no evidence to support the use of adjunctive technologies such as toluidine blue, brush biopsy or fluorescence imaging as a screening tool to reduce oral cancer mortality. A dye that stains a patch or a light that makes it fluoresce may help a clinician decide where to take tissue from. Neither replaces the tissue.

Is there a screening test, and who should have one?

The test is someone looking in your mouth, and the evidence for it depends on who is being looked at. The Cochrane review found that in a population screening trial, oral cancer mortality among tobacco or alcohol users fell by 24%, from 39.0 per 100 000 person-years in the control group to 30 per 100 000 in the screened group, a statistically significant difference. In the population as a whole there was no statistically significant difference, at 15.4 against 17.1 per 100 000 person-years. Screening worked where the risk was concentrated.

That evidence has real limits, and Cochrane says so: it rests on a single study with a high risk of bias that did not account for the effect of cluster randomisation in the analysis. Different bodies read this differently. The US Preventive Services Task Force concludes that current evidence is insufficient to assess the balance of benefits and harms of screening for oral cancer in asymptomatic adults. WHO's assessment of the evidence is that early detection programmes targeting high-risk populations, in selected settings where incidence is high such as South and South-East Asia and the Western Pacific islands, may reduce oral cancer incidence and mortality.

For an individual, the useful translation is this. If you use or have used tobacco in any form, areca nut or betel quid, or drink heavily, ask for your mouth to be examined at every dental visit and at least once a year, and do the mirror check yourself between visits. NIDCR notes the examination is painless and takes only a few minutes, covering the face, neck, lips, the entire mouth and the back of the throat. If none of those risks apply and your mouth is normal, a symptom lasting two weeks is still the signal that matters.

What does treatment involve, and why does early stage change everything?

In outline: surgery to remove the tumour, often with radiation therapy, with chemotherapy given after surgery in some cases and immunotherapy used in recurrent or metastatic disease. The US National Cancer Institute's treatment summary describes surgery as a common treatment for all stages of lip and oral cavity cancer, with early-stage disease often treated by wide local excision or radiation alone, while advanced disease may require much larger operations, sometimes removing part of the tongue and structures around it, combined with radiation.

That difference is the whole argument for the two-week rule. A small lesion can often be dealt with by one operation on one part of the mouth. A large one involves the tongue, the jaw or the neck, and the mouth is not a spare part: the lips and oral cavity are used for breathing, eating and talking, so the treatment itself costs speech, swallowing and appearance in a way that early treatment usually does not. People being treated often need specific help adjusting to those effects.

One more thing from the same summary, because it is actionable on the day of diagnosis: for patients who smoke, the chance of recovery is better if they stop smoking before beginning radiation therapy. Stopping is not only prevention. It also changes how well treatment works for someone who already has the disease, which is worth knowing for anyone who assumes the decision no longer matters once a diagnosis has been made.

A mouth problem: how fast does it need to be seen?

Routine — see a doctor

Make an appointment with a dentist or doctor within days, not months, for any mouth ulcer or sore that has not healed in two weeks, a white or red patch that cannot be wiped off, a thickened area in the cheek or lip, a persistent sore throat or hoarseness, or one-sided ear pain with no ear problem. Take a note of when you first noticed it. Repeat courses of antibiotics, vitamin tablets or mouth gels for a patch that keeps coming back are not a plan; ask specifically whether the lesion should be biopsied.

Same-day — call promptly

Ask for an urgent assessment or referral the same day for a new painless lump in the neck, especially alongside a mouth ulcer or patch that has already lasted more than two weeks, teeth becoming loose with no dental cause, numbness of the tongue or lip, a patch that is growing, ulcerating or bleeding, swelling of the jaw that suddenly makes dentures fit badly, or a mouth that has become noticeably harder to open. These are the features that point to disease that has gone beyond the surface lining, and the difference between this week and next month is measured in stage.

Emergency — act now

Go to emergency care now for difficulty breathing or noisy breathing, difficulty swallowing your own saliva or drooling because you cannot swallow, bleeding from the mouth that will not stop with pressure, or rapid swelling of the mouth, floor of the mouth or neck. These threaten the airway, and an airway problem is measured in minutes rather than days.