Can Seborrheic Keratosis Become Cancerous?

Seborrheic keratoses are benign growths, and the vast majority will never turn cancerous. But the relationship between these common skin lesions and cancer is more tangled than a simple “no” suggests. Rare cases of direct transformation into squamous cell carcinoma have been documented, certain subtypes carry higher risk profiles, and perhaps most important for the average person, melanoma and other skin cancers can look strikingly similar to a harmless seborrheic keratosis or even grow hidden underneath one.

What the Research Actually Shows About Transformation

The textbook line on seborrheic keratoses (SKs) has long been that they are entirely benign, full stop. And for practical purposes, that holds. But a small body of pathology research has found that squamous cell carcinoma can develop within an existing SK, particularly in older patients with large lesions on sun-exposed skin. An 11-case clinicopathologic study confirmed that SKs can undergo malignant transformation, with large size and older age as the key risk factors.¹ A review of the literature found that this in-situ transformation occurs more frequently in areas that get chronic sun exposure, though the overall association between SK and skin malignancies appears to be coincidental rather than causal.²

To put this in perspective, these documented cases remain extremely rare relative to the millions of SKs that exist on human skin worldwide. Most dermatologists will go through their entire career seeing only a handful, if any, cases of genuine transformation. The clinical significance is not that your SK is likely to become cancerous, but rather that the possibility is not zero and that certain features, particularly rapid growth, large size, and location on chronically sun-damaged skin in an elderly patient, deserve a closer look.

The Clonal Subtype Worth Knowing About

Not all seborrheic keratoses are created equal under the microscope. A variant called clonal (or nesting) seborrheic keratosis has drawn attention because of its more worrisome behavior. Clonal SKs feature distinct nests of cells within the epidermis that can look unsettlingly similar to Bowen’s disease, which is squamous cell carcinoma in situ. This resemblance is not just cosmetic. Research has found that incompletely removed clonal SKs with cellular atypia recur more frequently than those without atypia, and such lesions carry a meaningful risk of progressing to at least squamous cell carcinoma in situ upon recurrence.³

Distinguishing clonal SK from Bowen’s disease is genuinely difficult, even for experienced pathologists. A panel of immunohistochemical markers including CK10, Ki-67, and p16 has proven useful for telling them apart, since Bowen’s disease tends to show increased Ki-67 activity and strong p16 staining, while clonal SK shows a different pattern of CK10 loss.⁴ Even deep-learning algorithms trained to distinguish the two have struggled with irritated SKs and Bowen’s disease.⁵ The practical takeaway here is that if your dermatologist biopsies what appeared to be a routine SK and the pathology report mentions “clonal” features or atypia, a more thorough removal and closer follow-up make sense.

When Cancer Hides Behind or Inside an SK

A more common clinical concern than direct transformation is the collision tumor, where a skin cancer grows adjacent to or underneath a seborrheic keratosis. The SK essentially acts as a visual shield, masking the cancer from the patient and sometimes from the clinician. In a multicenter study by the International Dermoscopy Society looking at collision skin lesions, researchers found melanomas, basal cell carcinomas, and squamous cell carcinomas all appearing as part of collision lesions with seborrheic keratoses.⁶ One published case documented a basal cell carcinoma near the eye that was hidden by an overlying SK.⁷

This scenario matters because a patient or doctor who recognizes a lesion as “just a seborrheic keratosis” may not look further. A collision tumor is not the SK transforming; it is a cancer that happens to share the same real estate. The two processes are biologically independent but physically overlapping. This is one of the strongest arguments for paying attention when an SK changes appearance, develops an unusual border, or has areas that look different from the rest of the lesion.

Melanoma’s Seborrheic Keratosis Disguise

Perhaps the most clinically dangerous overlap between SKs and cancer is not transformation at all but mimicry. Melanoma can present with the waxy, stuck-on appearance that clinicians associate with harmless SKs. A cross-sectional study found that roughly 40% of melanomas that clinically resembled seborrheic keratoses were initially misclassified as benign.⁸ That is a startling misclassification rate for the deadliest common skin cancer.

Dermoscopy, the magnification tool dermatologists use to examine skin lesions, substantially improves the odds. In a study of 134 melanomas that clinically mimicked SKs, about 82% were correctly identified as melanoma when dermoscopy was used. Key features that tipped off the diagnosis included pigment network, blue-white veil, and globules and dots.⁹ Still, about 18% of those melanomas looked enough like genuine SKs to fool even dermoscopy.¹⁰ Those stubborn mimics are the cases that get caught only when a biopsy is performed for another reason, or when the lesion changes in a way that prompts a second look.

The clinical lesson here goes beyond academic interest. If you have a spot that “looks like a seborrheic keratosis” but is new, growing unevenly, has multiple colors, or just feels different from your other SKs, pushing for a biopsy is reasonable. The cost of removing a harmless growth is trivial compared to the cost of missing a melanoma.

Shared Mutations, Different Fates

One reason the SK-versus-cancer question keeps surfacing is that seborrheic keratoses carry some of the same genetic mutations found in actual cancers. The most common is in the FGFR3 gene, which regulates cell growth and was detected in about half of SKs in one genetic study. Mutations in PIK3CA, a gene involved in cell signaling that is also mutated in many cancers, were found in about a third of the same lesions. TERT promoter mutations, which affect cellular aging, appeared in about a quarter.¹¹ Family studies confirmed that FGFR3 and PIK3CA mutations arise within the skin cells themselves and are not inherited through the germline.¹²

These mutations accumulate in skin cells over time, driven in part by ultraviolet exposure. Research has shown that repeated UV exposure increases DNA damage and reactive oxygen species in keratinocytes through a specific enzyme pathway, contributing to the cellular changes seen in SK.¹³ Animal research has also demonstrated that UVB radiation can induce SK through enhancement of epidermal growth factor receptor expression, and that antioxidants can reduce these changes.¹⁴

So why do SKs stay benign despite harboring cancer-associated mutations? The current thinking is that these mutations alone are not enough. Cancer typically requires multiple hits across several control pathways. SKs seem to get one or two hits and then stop, remaining in a kind of controlled overgrowth without the additional damage needed for full malignant behavior. The mutations shared between SKs and certain cancers tell us something about how sun-damaged skin accumulates errors, but they do not mean the SK itself is precancerous in the way that, say, an actinic keratosis is.

The Sign of Leser-Trélat

There is one scenario where a sudden crop of seborrheic keratoses should raise a red flag, not because the SKs are becoming cancerous, but because they may signal a cancer elsewhere in the body. This is called the sign of Leser-Trélat: the sudden eruption of many SKs, or the rapid growth of existing ones, as a paraneoplastic phenomenon. A systematic review found that about 76% of patients with eruptive seborrheic keratoses had a co-occurring malignancy. In those cases, the eruptive SKs preceded the cancer diagnosis with an average lead time of four months. The cancers most commonly associated were cutaneous T-cell lymphoma and gastrointestinal adenocarcinoma.¹⁵

The sign of Leser-Trélat is rare enough that most dermatologists consider it a clinical curiosity, and some debate whether it is a true paraneoplastic sign or a statistical coincidence given how common both SKs and cancer are in older adults. Still, the systematic review data is hard to ignore. If you notice dozens of new SKs appearing over a short period, or your existing ones suddenly start growing or getting irritated, it is worth mentioning to your doctor, particularly if you have other unexplained symptoms like weight loss or fatigue.

How Common Are SKs in the First Place

Part of the reason seborrheic keratoses generate so much clinical traffic is their sheer ubiquity. A study of people aged 15 to 30 found that almost a quarter of them already had at least one SK, with prevalence rising from about 16% in teenagers to about 32% in people aged 25 to 30. The chance of having an SK increased by roughly 62% with every five years of age.¹⁶ By middle age, most people have at least a few. By old age, most people have many. Independent risk factors include advancing age, genetic predisposition, and cumulative ultraviolet light exposure, with the FGFR3 gene mutation pathway implicated as a key driver.¹⁷

This prevalence explains why even a vanishingly small transformation rate translates to a noticeable number of case reports in the literature. When billions of SKs exist across the world’s population, even a one-in-a-million event generates a substantial medical record. It also explains the economic burden of SK management: an older analysis of Medicare data found that nondermatologists generated about $9 million in costs from high-intensity SK management, compared to $6 million by dermatologists, suggesting that familiarity with the diagnosis matters for avoiding unnecessary procedures.¹⁸

When a Biopsy Makes Sense

Current clinical guidelines say that SKs should be diagnosed clinically and generally do not need treatment. However, a biopsy is reasonable when a lesion is suspicious, particularly if it is bleeding, inflamed, or changing in a way that departs from a typical SK pattern. If removal is desired, curettage or cautery works well for raised lesions and cryotherapy for flat ones. Full excision is usually avoided because it is more invasive and leaves a worse scar for what is almost always a benign growth.¹⁹

The situations that should prompt you to ask for a biopsy include:

• Rapid change: An SK that suddenly grows, darkens, or develops irregular borders.
• Unusual appearance: A lesion your doctor initially calls an SK but that has areas of different colors, particularly black, blue, or red zones, or lacks the classic waxy texture.
• Bleeding or ulceration: While SKs can occasionally bleed from trauma like catching on clothing, spontaneous bleeding or persistent ulceration is not typical.
• Eruptive pattern: Many new SKs appearing over weeks to months, as noted in the Leser-Trélat discussion above.
• Solitary atypical lesion: A single lesion in a sun-exposed area on an elderly patient that looks somewhat like an SK but does not quite fit the pattern of the person’s other SKs.

In practice, many SKs do get biopsied simply because they cannot be confidently distinguished from something more concerning based on appearance alone. That is not a failure of the system; it is the system working as intended. The threshold for biopsy should be low when there is any doubt, because the procedure is minor and the downside of missing a skin cancer is severe.

Emerging Tools for Sorting SKs From Cancers

The challenge of distinguishing SKs from skin cancers has driven interest in non-invasive imaging and artificial intelligence. Reflectance confocal microscopy, which lets clinicians view skin structure at near-cellular resolution without cutting, has shown promise. In a study of 45 SKs, characteristic features like epidermal projections and keratin-filled invaginations were present in the vast majority of cases, and 93% of biopsied SKs displayed at least three characteristic benign findings with no features suggestive of malignancy.²⁰ This kind of imaging could eventually reduce unnecessary biopsies by providing near-biopsy-level confidence without a blade.

Smartphone-based AI is further behind. One neural network model designed to classify melanomas, nevi, and SKs from phone photos achieved high sensitivity for melanomas but only about 52% sensitivity for seborrheic keratoses, meaning it missed roughly half of them.²¹ The researchers attributed this to insufficient training data. For now, a smartphone app might reassure you that a melanoma is not being missed, but it is unreliable at confirming that something is a harmless SK. The technology is improving fast, but it is not yet a substitute for a dermatologist’s trained eye and a dermoscope.

The HPV Connection

A recurring question in the dermatology literature is whether human papillomavirus plays a role in SK development. The connection is clearest in patients with a rare inherited immune condition called epidermodysplasia verruciformis, where HPV has been detected in a large proportion of SK lesions.²² In the general population, the evidence for HPV as a driver of typical SKs is weaker and inconsistent. Various studies have detected HPV DNA in SK tissue, but detection does not prove causation; the virus could simply be a passenger in the skin rather than a growth driver. This remains an open research question, and at this point there is no clinical recommendation to treat or monitor SKs differently based on HPV status.