Actinic keratosis: what therapy actually involves

No — and it is not nothing, either. It is a precancer: a patch of cells carrying sun-induced mutations that have started growing abnormally but have not become invasive. The risk of any one patch turning into squamous cell carcinoma in a given year is small. The catch is that these lesions rarely come singly and the risk runs for life, which is why the standard advice is to have them assessed and usually treated rather than to watch and hope.

Sometimes — individual lesions do regress spontaneously, which sounds more reassuring than it is. Regressed lesions can return, the field that produced them remains, and nobody can tell by looking which lesion will fade, which will persist, and which will progress. Clinicians treat them because treatment is quick and low-risk while the untreated worst case is a skin cancer; that asymmetry, not alarm, is the logic.

No — and this one is worth being blunt about. Exfoliants smooth off the scale, which feels like progress and re-forms within weeks, because the abnormal cells sit deeper than any cosmetic product reaches. Nothing sold without a prescription is approved to treat actinic keratosis anywhere. A rough patch that keeps returning after exfoliation is behaving exactly like an actinic keratosis, and that pattern is a reason to book an appointment, not to buy a stronger acid.

Because those are answers to two different questions. Freezing removes the discrete lesions you can point to. The cream — fluorouracil and its relatives — treats the whole field of sun-damaged skin, including early lesions too small to see, which is why it is prescribed by area rather than by spot. On skin with many lesions, treating one spot at a time becomes whack-a-mole; the field approach resets the whole board.

Almost certainly working — fluorouracil and imiquimod clear damaged cells by provoking inflammation against them, so a brisk red, crusted reaction in the treated area is the mechanism visible on the surface. It is also self-limited, settling within a couple of weeks of finishing. That said, 'expected' has a range: if the reaction seems beyond what you were told, is spreading past the treated area, or you were not warned at all, contact the prescriber rather than pushing through on faith.

It has real evidence in the right group — a phase 3 trial in people with previous skin cancers found that nicotinamide 500 mg twice daily reduced new keratinocyte cancers and actinic keratoses over a year. Two caveats keep it a conversation rather than a recommendation: the benefit is established in that high-risk group, not in everyone with a first lesion, and a later trial in transplant recipients did not show the same effect. It is cheap and over-the-counter, which makes it tempting to self-prescribe — but whether it earns a place in your particular situation is a question for the clinician who knows your history.

You cannot fully — the ultraviolet exposure that caused them is already banked in the skin, and the field keeps expressing it. What you control is the rate and the stakes: daily broad-spectrum sunscreen on the exposed territory measurably slows new lesions, a hat does more for a thinning scalp than any product, tanning beds are permanently off the table, and a standing schedule of skin checks turns 'more will come' from a threat into a managed expectation. Prevention here is unglamorous and genuinely effective, which is a rarer combination than it sounds.

For years the honest answer was 'probably, but nobody has shown it directly'. Then a randomised trial in 932 people with a history of skin cancer gave a single two-to-four-week course of fluorouracil cream to the face and ears: in the following year, 1% of the treated group needed surgery for squamous cell carcinoma against 4% of the control group. The effect faded after that first year and basal cell carcinoma was unaffected — so it is a real benefit with a shelf life, not a permanent settlement, which is also a fair description of this whole condition.

The 2019 head-to-head compared fluorouracil, imiquimod, photodynamic therapy and ingenol mebutate gel. The fourth is no longer available: the European Medicines Agency suspended it in January 2020 and the marketing authorisation was withdrawn the following month, after a study found more skin cancers — squamous cell carcinoma in particular — in areas treated with it than in areas treated with imiquimod. The UK regulator suspended it on the same grounds. If you used it in the past, the advice is to mention that at a skin check and to report unusual growths in the treated area rather than to worry about it in the abstract.

Dry skin improves with moisturiser and moves around. A patch that re-forms in the same spot on sun-exposed skin for months is behaving differently.

Dry skin is diffuse, seasonal and responsive: it covers an area rather than a spot, it worsens with cold and hot showers, and a decent emollient changes it within days. The pattern that should make you book an appointment is narrower than that. One patch, in one place, on skin that has caught decades of sun, that flakes off and re-forms in exactly the same outline, and that moisturiser makes softer without making it go away. That persistence is the signature. Many people spend a year quietly exfoliating such a patch, because scrubbing does briefly smooth it, which reads as progress. The scale shed is only the surface; the abnormal cells sit deeper than any cosmetic product reaches, and they rebuild the scale within weeks. If you have been treating the same square centimetre of forehead, ear, forearm or bald scalp with a different product every few months for a year, the useful next step is not another product.

Cumulative ultraviolet exposure damaging the DNA of skin cells, classically the p53 gene that normally halts a damaged cell before it divides.

Every hour of ultraviolet exposure leaves a small amount of DNA damage in keratinocytes, the cells that build the skin's outer layer. Most of it is repaired. A fraction is not, and over decades the unrepaired damage concentrates in genes that control cell division — p53 above all, which normally acts as the brake that stops a damaged cell replicating. When enough of those brakes fail in one patch of skin, the cells there start dividing abnormally and producing the rough, adherent scale you can feel. That is an actinic keratosis: not an infection, not an allergy, not a reaction to anything you applied, and not a consequence of anything happening in your body this year. It is a record of light. The corollary is worth stating, because it shapes everything else: the exposure that produced this patch was never limited to this patch. The whole surface that caught that sun carries the same history, which is why these lesions arrive in fields rather than singly, and why an appointment about one spot becomes an examination of everything.

No — most individual lesions never progress, with commonly quoted transformation figures well under ten per cent; the risk lies in having many, for many years.

Precancer describes what the cells are, not what they are destined to do. An actinic keratosis contains keratinocytes carrying the mutations squamous cell carcinoma grows from, but the great majority never take that step, and some regress on their own. Patient-facing sources commonly cite transformation of individual lesions at somewhere between five and ten per cent. Set against that is the fact these lesions rarely arrive singly and the risk runs for the rest of your life, so the meaningful figure is a small annual probability multiplied across a field of lesions and a couple of decades. That multiplication is the reason clinicians treat rather than watch: treatment is quick and low-risk, while the untreated worst case is a cancer. So the honest framing is neither of the two you will hear. It is not a countdown, and it is not nothing. It is a manageable risk that becomes properly manageable once someone has looked at all of your skin rather than the one spot.

No. Every medicine that clears these lesions is prescription-only, and no over-the-counter product may legally claim to treat actinic keratosis in the United States.

This is unusual on a skincare site, and it is worth stating without hedging. There is no over-the-counter tier to try first here. Fluorouracil, imiquimod, tirbanibulin and prescription-strength diclofenac for this indication are prescription drugs everywhere, and the procedures are performed in clinic. That line is not editorial caution — no over-the-counter drug monograph covers actinic keratosis, so a product making the claim is making it illegally. The regulatory logic is worth understanding, because it is also the clinical logic: a lesion that might already be an early squamous cell carcinoma needs to be examined before anything is applied to it, and the worst outcome in this condition is not a treatment that fails but a cancer treated as though it were dry skin. What the aisle can do is real but different. Daily sunscreen has randomised evidence for reducing new lesions, an SPF lip balm protects the site everyone forgets, and an emollient makes a rough patch comfortable while you wait for the appointment that actually treats it.

Not until roughly four to eight weeks after the course ends. The treated skin has to finish reacting and re-heal before anyone can judge what cleared.

During a field course and for a week or two afterwards, the area is red, crusted and inflamed, and nothing useful can be read from it — the lesions are indistinguishable from the reaction around them. Healing takes one to two weeks after the last application, and the skin then continues settling for several weeks more. Clinicians typically assess clearance somewhere between four and twelve weeks after the course, and many photograph the field before starting so there is something objective to compare against. Judge it against that appointment rather than against the mirror in week three. What you will usually see when it settles is skin that is smoother overall, with a few stubborn spots remaining — that is a common outcome rather than a failure, and those spots are then handled individually. A lesion still present at the review, or one that re-forms promptly in a treated site, is the specific finding to report, because failure to clear sits on the standard list of reasons to biopsy rather than re-treat.

Redness, scaling and crusting in the treated area are the mechanism. Reaction spreading well beyond it, severe swelling, or pain you cannot manage warrants a call.

Fluorouracil clears damaged cells by provoking inflammation against them, so a brisk red, raw, crusted response in the treated field is the drug working rather than an allergy — and it is often far more dramatic than people expect. The normal version is confined to where you applied it, peaks in weeks two and three, and settles within a couple of weeks of stopping. The versions worth a phone call are different in kind rather than degree: reaction spreading well outside the treated area, marked swelling, pain that interferes with sleep or eating, or a treated site that becomes increasingly painful with pus and spreading warmth, which suggests infection rather than the expected reaction. Hospital patient information for this treatment says plainly that if the skin becomes excessively inflamed you can stop before the end of the course and let it settle, sometimes restarting at reduced frequency. That is a decision to make with the prescriber, which is exactly why you should have asked, before starting, what their threshold looks like.

Mechanism and schedule mostly. Fluorouracil kills damaged cells directly, imiquimod recruits your immune system, tirbanibulin blocks cell division over a five-day course.

Fluorouracil is the workhorse: two to four weeks of application, the fiercest reaction, and the strongest evidence for clearance. Imiquimod does something different — it activates a local immune response against damaged cells, over a schedule measured in weeks to months depending on the regimen, with its own inflammatory reaction and occasional flu-like days. That difference shows up late: in a three-arm randomised comparison, imiquimod had the highest sustained clearance a year after treatment and the cosmetic outcomes investigators rated best, despite clearing fewer lesions at the end of the course than fluorouracil. Tirbanibulin is the newest, a five-day course with a generally milder reaction and a much higher price. Prescription-strength diclofenac gel is the slow, gentle option run over two to three months with more modest clearance, now used mainly in sequence rather than alone. Which one you are offered depends on the site, how much damage there is, what you can tolerate and what is affordable — and most people with ongoing lesions meet more than one over the years.

Medical, not cosmetic. Actinic keratosis is a precancerous condition, and treatment is generally covered as medical care in insurance systems.

This trips people up more than it should, because the treatments overlap with things sold cosmetically — freezing, peels, lasers — and because the lesions look like a beauty complaint. They are not classified that way. A precancerous diagnosis makes destruction of lesions and prescribed field treatment medical care, and in insurance-based systems it is generally covered, with the usual caveats about deductibles, copays and network. Coverage can get narrower at the edges: newer branded agents may need prior authorisation or step therapy through a generic first, photodynamic therapy involves clinic time that is billed and covered variably, and cosmetic resurfacing done partly for appearance sits in a different category. The reason this matters practically is that reluctance to spend money on skin stuff keeps some people from treatment their coverage already includes. If cost is the barrier, ask the office to check coverage before the appointment rather than assuming, and ask specifically whether a generic route avoids a prior-authorisation delay.

No. Fluorouracil is contraindicated in pregnancy, and field creams generally are avoided — treatment is postponed rather than adapted.

Take this as a firm position rather than a cautious one: fluorouracil should not be used during pregnancy, and the American Academy of Dermatology lists it as contraindicated for this indication. The other field agents are avoided as a group, not because each carries a specific documented harm but because there is no situation in this condition that requires accepting an unknown. That reasoning is worth understanding, because it is what makes the answer easy. Actinic keratosis progresses over years to decades. Deferring elective field treatment for the length of a pregnancy costs almost nothing against that timescale. So tell whoever is prescribing that you are pregnant, or planning to be, before a course is written — this is one of the few places on this site where the timing of the disclosure genuinely changes the plan. Lesion-directed options fill the gap in the meantime, and a review after the pregnancy should be agreed in advance so deferral does not quietly become abandonment.

Often in sun-protected places — palms, soles, under nails, inside the mouth, the buttock or genital area — which is exactly where nobody thinks to look.

The American Academy of Dermatology is direct about this: in people with darker skin tones, skin cancer frequently develops on the palms, the soles of the feet, fingers and toes, the nails, inside the mouth or on the lip, and in the buttock, anal or genital area. Those are not the sites a sun-damage narrative points you towards, which is a large part of why diagnosis is often late and the cancer already advanced by the time it is found. That changes what a self-check should look like. Include the soles and between the toes, the palms, the nail beds — looking for a dark line or streak — the lower lip and the inside of the mouth, and any area that has been injured or scarred in the past. Look for sores that will not heal or keep coming back, patches that are growing or changing, areas of darkening, and nail changes. Any of those warrants examination regardless of how much sun the site has had.

Physically yes, and most people do. The reaction on a face is conspicuous for roughly two weeks, which is why courses get scheduled around work rather than hidden.

Be realistic about what the middle of a fluorouracil course looks like: bright red, scaly, crusted skin across a whole forehead, cheek or scalp, at a level people comment on. That is the drug working, and it settles within a week or two of the last application, but nothing available conceals it convincingly. Three approaches actually help. Schedule the course into a quieter fortnight — winter, annual leave, a stretch without client-facing work — since prescribers do this routinely and will work backwards from a date if you give them one. Tell colleagues briefly what it is, because a one-line explanation prevents a fortnight of speculation and is far less exhausting than deflecting. And if visible reaction is genuinely incompatible with your job, say so before the prescription is written: the shorter tirbanibulin course, a slower diclofenac regimen, or lesion-directed treatment in clinic are all legitimate alternatives chosen for exactly this reason. Stopping mid-course because of how it looks is the outcome worth avoiding.

It changes the question. Recurrence in a treated site is a standard trigger for biopsy, because some lesions that behave this way are already squamous cell carcinoma.

There is a meaningful difference between a new lesion appearing nearby — the field doing what fields do — and the same lesion re-forming in the footprint of one that was treated. The second is on the standard list of reasons to take tissue rather than repeat the treatment, alongside thickening, tenderness, bleeding and rapid growth. The reason is uncomfortable but simple: a proportion of lesions that look and feel like actinic keratoses turn out to be invasive squamous cell carcinoma under the microscope, and the ones that resist adequate treatment are enriched for exactly that. In a series of clinically diagnosed lesions sent for pathology, roughly one in twenty-five was an occult squamous cell carcinoma, and clinical recognition performed worse still in people with a previous skin cancer. So when you go back, use the specific words: it recurred in the treated site. That phrasing is what moves the conversation from another freeze to a biopsy, and the great majority of those results come back reassuring.

Growth, thickening, tenderness, bleeding, ulceration, a spot that will not heal, or new lesions multiplying while you are on immunosuppressants.

The routine version of this condition is a stable rough patch that has been there for months and behaves the same way every time you check. What earns a faster appointment is change. The specific features to act on are a patch that is growing or becoming raised, one that has thickened or feels firm and indurated, one that has become tender or painful, one that bleeds or ulcerates, and one that simply never heals. Rapid multiplication of new lesions while taking immunosuppressant medication belongs in the same category. Say the feature out loud when you book — waiting lists are triaged on exactly those words, and a lesion that has changed is not a routine referral. None of these features means cancer; they mean the lesion is behaving in a way that needs a trained look rather than another month of watching, and the answer is often a small biopsy with a lot of information in it. If a lesion is bleeding and not healing at all, do not wait for a routine slot.

Non-ablative fractional lasers improve how the skin looks without clearing the lesions. In one study, biopsies after five treatments still showed actinic keratosis or worse.

This is the most instructive negative result in the whole condition, and it is worth knowing before spending money. Non-ablative fractional resurfacing improves texture, tone and general photodamage — the appearance genuinely gets better. In a study following facial lesions with histology after five treatments, virtually every biopsy still showed actinic keratosis or something worse. Clinical improvement without histological clearance is precisely the failure that matters in a precancer, because the endpoint here is not how the skin looks but whether the abnormal cells are gone. Skin that looks treated and is not is arguably worse than skin that looks untreated, since it removes the visual cue that would have prompted an appointment. Ablative resurfacing is a different proposition — it removes the epidermis outright, has held recurrences off for years in follow-up series, and beat a fluorouracil course on recurrence at twelve months in one comparison — at the cost of substantial downtime and a recognised risk of permanent hypopigmentation.

Everything moves one notch up in urgency. Lesions arrive faster, progress more readily, and surveillance shifts from occasional to scheduled — often every six to twelve months.

Long-term immunosuppression rewrites this condition's timetable. Transplant recipients in particular develop these lesions earlier and in greater numbers, and progress to squamous cell carcinoma far more readily — the risk of squamous cell carcinoma in organ transplant recipients is many times that of the general population, which is why transplant services build dermatology surveillance into routine follow-up rather than waiting to be asked. European and US guidance points towards specialist full-skin examination every six to twelve months for this group, with intervals adjusted by risk. The treatment tiers themselves do not change; what changes is how early they are reached for, how often they are repeated, and how quickly an odd lesion gets biopsied instead of watched. Two specifics worth carrying: a lesion that fails to clear is investigated rather than re-treated, and oral nicotinamide does not carry over automatically, since a randomised trial in transplant recipients did not show the benefit seen in immunocompetent people.

Not for these lesions. No over-the-counter product may legally claim to treat actinic keratosis, and anything marketed that way is selling exfoliation at best.

Search results for these lesions fill up with products promising to dissolve, remove or clear sun spots, and the category is worth reading with a specific lens. Every medicine that clears an actinic keratosis is prescription-only, and no over-the-counter drug monograph covers this condition — so a product claiming to treat it is making a claim it is not permitted to make, which tells you something about the rest of the label. What is usually inside is an exfoliating acid, sometimes a retinol, occasionally an escharotic paste that will burn a hole in your skin. The first two smooth the surface temporarily; the third is genuinely dangerous. Watch for the tells: before-and-after photographs, the word precancer used loosely, testimonials rather than trials, and any instruction to expect the lesion to scab and fall out. The money is better spent on a sunscreen you will use daily, which is the one aisle product with randomised evidence behind it in this condition.