Keratoacanthoma vs Squamous Cell Carcinoma: Distinguishing Traits

Keratoacanthoma and squamous cell carcinoma can look so similar under a microscope that even experienced dermatopathologists disagree about what they are seeing, and whether these two diagnoses truly represent separate diseases remains one of the most persistent debates in skin pathology.¹ Despite that uncertainty, the two behave very differently in the body. What separates them clinically, under a dermatoscope, and at the molecular level is worth understanding because those differences shape real treatment decisions.

Why the Distinction Matters and Why It Is So Hard

Keratoacanthoma grows fast, often ballooning to a centimeter or more in just a few weeks on sun-exposed skin. It typically forms a dome-shaped nodule with a central keratin-filled crater, looking a bit like a tiny volcano. Many KAs then do something unusual for a skin tumor: they stop growing and shrink on their own over the course of several months, sometimes disappearing entirely. Squamous cell carcinoma, by contrast, tends to grow more slowly, lacks that stereotypical crateriform shape, and does not regress. Left untreated, SCC can invade deeper tissue and, in a small fraction of cases, spread to lymph nodes or distant organs.

The problem is that these behavioral differences only become clear over time. At a single clinical snapshot, a rapidly growing KA and an aggressive SCC can be nearly indistinguishable. Even on biopsy, the two share so many histological features that no reliable single marker has been found to separate them definitively.¹ This overlap is the reason most clinicians treat a suspected KA the same way they would treat SCC: they cut it out. Waiting around to see whether a lesion regresses is a gamble, because if the diagnosis turns out to be wrong, the patient has lost valuable time.

How Spontaneous Regression Works

The hallmark trait of a true keratoacanthoma is its ability to self-resolve. Research using a mouse model that mimics human KA behavior has shown that regressing tumors shift from active growth into a differentiation program. Wnt signaling, a pathway the body normally uses during hair follicle regeneration, gets hijacked to sustain tumor growth. When the retinoic acid signaling pathway kicks in and suppresses Wnt, the tumor starts to differentiate and ultimately regresses.²

There is also strong evidence that the immune system plays a central role. Compared to SCCs, keratoacanthomas harbor significantly more CD3-positive and CD4-positive T cells invading the tumor itself. Those immune cells in KAs also show greater signs of activation, expressing higher levels of the interleukin-2 receptor. The tumor cells in KAs, meanwhile, display more of the adhesion molecule ICAM-1 and the differentiation marker keratin 10, which together suggest the tumor is essentially cooperating with the immune system in its own dismantling.³

That immune involvement matters for understanding why KAs behave differently in people whose immune systems are suppressed, a topic discussed further below.

The Immune Microenvironment as a Distinguishing Feature

More recent work has refined the picture. When researchers compared the T-cell profiles of KAs, invasive SCCs, and precancerous actinic keratoses, they found that KAs had significantly more CD4-positive and FOXP3-positive cells but fewer CD8-positive cells compared with invasive SCC. Invasive SCCs, in turn, expressed significantly more PD-L1, the “don’t eat me” signal that tumors use to evade immune attack.⁴

This contrast is revealing. KAs seem to invite an immune response that ultimately destroys them, while SCCs develop molecular escape strategies like PD-L1 upregulation. It is one of the more compelling pieces of evidence that these tumors have fundamentally different relationships with the body’s defenses, even when they look similar under a standard microscope.

Molecular and Genetic Markers That Separate the Two

Because conventional microscopy often fails to provide a definitive answer, researchers have searched for molecular markers that might distinguish KA from SCC on a biopsy slide. Several have shown promise. Telomerase activity, a marker of cell immortality, was detected at higher levels in about two-thirds of SCCs compared with fewer than one in five KAs. Similarly, strong expression of the tumor-suppressor protein p53 and the inflammatory enzyme COX-2 was found in roughly 70% and 65% of SCCs, respectively, but in fewer than one in ten KAs.⁵ These differences were statistically robust and support the view that KA and SCC are biologically distinct, not just stages of one disease.

At the chromosomal level, the gap widens further. One study using comparative genomic hybridization found that about 84% of SCCs carried genomic aberrations compared with roughly 37% of KAs, pointing to far greater chromosomal instability in SCC. Specific gains on chromosomes 1p, 14q, 16q, and 20q, along with losses from 4p, appeared significantly more often in SCCs, while loss from 9p was more characteristic of KAs.⁶ A separate array-based study of 98 KAs and 22 SCCs confirmed that the distribution of genetic aberrations differed significantly between the two tumor types, with unsupervised clustering separating them into distinct groups.⁷

These genomic studies are some of the strongest arguments that KA is its own entity rather than a well-differentiated variant of SCC, though critics note that biopsy-classified KAs could include misdiagnosed SCCs, which would muddy the results. The debate has not settled.

What a Dermatoscope Can and Cannot Show

Dermoscopy, the handheld magnification tool that dermatologists use during skin exams, can help distinguish KA and SCC from other raised, non-pigmented skin lesions, though it is less reliable at separating the two from each other. In a study evaluating over 200 raised nonpigmented lesions, white circles around follicular openings had the highest specificity (87%) for identifying SCC and KA as a combined group, while keratin had the highest sensitivity (79%). Central keratin was more common in KA than in SCC (about half of KAs versus 30% of SCCs).⁸

A separate study looking at vascular patterns found a couple of useful clues. Central keratin plugs were visible in the majority of KAs but only about 13% of invasive SCCs. Ulceration was far more common in SCC, appearing in 80% of those lesions versus roughly 43% of KAs. And a white background color was more associated with KAs, while a mixed color background was more typical of invasive SCCs.⁹ Both linear irregular vessels and a peripheral vascular pattern showed up in KAs and SCCs at comparable rates, so vessels alone do not reliably separate them.

In practice, dermoscopy helps a dermatologist decide that something is suspicious enough to biopsy, but the final distinction usually requires tissue under a microscope, sometimes supplemented by the molecular markers described earlier.

Treatment Approaches and the “Just Cut It Out” Philosophy

Given the diagnostic uncertainty, most clinicians default to excision. In a review of nearly 1,000 published KA cases, only about 20% were ever observed through to spontaneous regression, and fewer than 2% were documented with serial photographs confirming true self-resolution.¹ Excision has the obvious advantage of producing a complete specimen that a pathologist can evaluate for depth of invasion and clear margins. For a true KA, surgery is curative. For an SCC masquerading as a KA, early excision prevents potential complications.

When surgery is not ideal, perhaps because of the lesion’s size, location on the face, or the patient’s medical status, non-surgical options exist. Intralesional injections of 5-fluorouracil or methotrexate can be effective. A systematic comparison found that both topical and intralesional treatments achieved high eradication rates (92% to 100%), with intralesional 5-fluorouracil leading to faster healing than intralesional methotrexate (about 3.7 versus 4.6 weeks).¹⁰ A smaller trial comparing the two intralesional agents head-to-head found complete clearance in 80% of the 5-FU group and 70% of the methotrexate group, with 5-FU needing fewer sessions.¹¹ These non-surgical routes are usually reserved for confirmed or strongly suspected KAs; if there is any doubt that the lesion might be SCC, excision remains the safer path.

When Keratoacanthomas Do Not Behave

One reason the debate stays unresolved is that some lesions diagnosed as KAs go on to behave aggressively, invading deeply or recurring after removal. Case reports of initially diagnosed “giant keratoacanthomas” that turned out to be invasive squamous cell carcinomas are not uncommon in the literature.¹² Whether these cases represent true KAs that transformed into SCCs, or were simply SCCs that were misdiagnosed from the start, is the crux of the taxonomic disagreement. Some dermatopathologists argue that any KA capable of aggressive behavior was never really a KA, while others maintain that KA sits on a biological continuum with SCC and that transformation is possible.

Immunosuppressed patients are a particularly important group here. Organ transplant recipients, for example, have a higher incidence of both KAs and SCCs. Reports of KAs in these patients undergoing documented malignant transformation and recurring after excision suggest that a weakened immune system removes the very mechanism that normally drives KA regression.¹³ For immunosuppressed patients, the conservative “watch and wait” approach is essentially off the table, and aggressive treatment of any KA-like lesion is standard practice.¹⁴

Drug-Induced Keratoacanthomas

An interesting wrinkle emerged with the advent of BRAF-inhibitor therapy for melanoma. Drugs like vemurafenib and the broader-spectrum RAF inhibitor sorafenib were found to trigger the development of both SCCs and KAs as a side effect. The proposed mechanism involves paradoxical activation: in cells that carry normal (wild-type) BRAF, pharmacologic blockade of RAF redirects signaling through CRAF, which amplifies MAPK pathway activity and can push keratinocytes toward tumor formation.¹⁵

Drug-induced KAs tend to appear in clusters and often arise within weeks of starting therapy. For oncologists managing melanoma patients on BRAF inhibitors, frequent skin surveillance is now standard, and any suspicious lesion gets biopsied promptly. The phenomenon also underscores that KAs and SCCs share upstream signaling pathways, even if their downstream biology diverges.

Syndromic Settings and Multiple Keratoacanthomas

Most KAs occur as solitary lesions in older adults on sun-damaged skin. But multiple KAs can be a red flag for certain inherited conditions. Muir-Torre syndrome, an autosomal-dominant disorder caused by mismatch repair gene mutations, is characterized by sebaceous gland tumors or keratoacanthomas alongside internal malignancies, particularly colorectal and endometrial cancers.¹⁶ In these patients, the appearance of multiple KAs can be the first clinical sign that triggers workup for visceral tumors.

Xeroderma pigmentosum, a rare disorder of defective DNA repair, also produces KAs alongside a spectrum of sun-induced skin cancers. A comparison of the two syndromes found that KAs in xeroderma pigmentosum patients appeared much earlier in life (average age around 11 years) compared with Muir-Torre patients (average around 50 years), reflecting the different underlying genetic defects.¹⁷ In both syndromes, multiple KAs were the norm. Anyone presenting with more than one KA, especially at a young age, deserves investigation for an underlying genetic predisposition.

The Role of UV Exposure and HPV

Both KA and SCC share ultraviolet radiation as a major environmental risk factor. They appear overwhelmingly on sun-exposed sites like the face, ears, forearms, and hands, and are more common in fair-skinned people with cumulative sun damage. This shared etiology is another reason separating the two is so difficult: they arise from the same cell type (keratinocytes) under the same environmental pressure.

Human papillomavirus has been investigated as a potential contributor to both lesions, but the evidence is thin. A study examining HPV DNA in KA and SCC tissue samples found relatively low detection rates in both, concluding that HPV might be involved in only a subset of cases rather than playing a central role.¹⁸ This contrasts with mucosal SCCs (like cervical cancer), where HPV is the dominant driver. For cutaneous KA and SCC, sunlight remains the primary environmental culprit.

How Clinicians Navigate the Uncertainty in Practice

Given everything above, the practical reality is that many pathology reports will describe a lesion as “keratoacanthoma-type squamous cell carcinoma” or “well-differentiated SCC, keratoacanthoma type” rather than making a clean call. This hybrid language reflects genuine diagnostic humility. The label serves as a signal that the lesion looks more KA-like, may carry a better prognosis, but should still be managed with the same margins and follow-up as SCC.

For you as a patient, the key takeaway is that a rapidly growing dome-shaped nodule on sun-exposed skin needs prompt evaluation regardless of whether your dermatologist suspects KA or SCC. If a biopsy comes back as keratoacanthoma, your doctor will likely still recommend complete removal. The exception might be a small, classic-appearing KA in a cosmetically sensitive area where an intralesional injection is preferred, but even then, close follow-up is essential to confirm the lesion resolves. If you are immunosuppressed, on BRAF-inhibitor therapy, or have a history of multiple KAs, the threshold for aggressive treatment drops further. The diagnostic gray zone between KA and SCC is a problem for researchers and pathologists to sort out, but it need not translate into uncertainty about what to do: when in doubt, treat it.