You are on page 1of 18

Chapter 113

Epithelial Precancerous
Karynne O. Duncan, John K. Geisse, &
David J. Leffell


Unusual variants of actinic keratoses (AK) include
pigmented AK, spreading pigmented AK, proliferative AK, and conjunctival AK. The pigmented AK
presents as a flat, tan to brown, scaly papule or
plaque. Clinically, and many times dermatoscopically, it can be difficult to distinguish from a solar
lentigo or lentigo maligna.22 Spreading pigmented
AK is a particular lesion that is often found on the
face or scalp. It is typically a large lesion, often
reaching more than 1 cm in diameter, and presents
as a scaly, verrucous, or smooth plaque with variable colors. Proliferative AK is seen as an expanding,
red, oval, scaly plaque that reaches up to 4 cm in diameter. It has poorly defined borders. Conjunctival
AK is classified as a type of pinguecula or pterygium.
It appears as a wedge-shaped, opaque thickening
near the limbus, and it may extend onto the cornea
from the scleral conjunctiva.

The typical erythematous AK has characteristic
architectural and cytologic histopathologic features. All of the abnormalities are confined to the
epidermis, although dermal solar elastosis is usually
present and an inflammatory dermal infiltrate may
also be seen. The classic histopathologic findings
include foci of atypical, pleomorphic keratinocytes along the basal cell layer and protruding as
buds into the papillary dermis. The basal cell layer
appears more basophilic because of crowding of
the atypical keratinocytes. Overlying these foci of
abnormal cells, one often sees irregular acanthosis,
hyperkeratosis, and parakeratosis. There is notable
sparing of the adnexal epithelium, with orthokeratosis overlying these structures, which gives rise to
the characteristic pattern of alternating orthokeratosis and parakeratosis (see Fig. 113-3B). Cytologically, there is an increased nuclearcytoplasmic

ratio in the atypical keratinocytes in addition to

nuclear pleomorphism and mitoses.
Just as there are clinical variants of AKs, there are
histopathologic variants. These histopathologic
variants all share the classic features of the AK as
described earlier, but with some additional findings.
In HAK, the hypertrophic areas consist of marked
hyperkeratosis and parakeratosis. A cutaneous horn
that represents AK demonstrates changes similar
to but more exaggerated than those in HAKs with
massive tiers of hyperkeratosis and parakeratosis.
The base of such a cutaneous horn shows the typical histopathologic changes of an AK. An atrophic
AK has very mild hyperkeratosis and a thinned
epidermis that is devoid of rete ridges. Bowenoid
AK is a histopathologic variant that demonstrates
near full-thickness epidermal dysplasia, but unlike
true Bowen disease (BD) or SCC in situ, there is no
or very little involvement of the adnexal epithelium.
In proliferative AK there are exaggerated bud-like
downgrowth of atypical keratinocytes into the
papillary dermis. Although difficult to distinguish
from SCC at times, these proliferative bud-like
downgrowth are contiguous with the epidermis in
the proliferative AK, unlike in true SCC.
In the pigmented AK one sees excessive amounts
of melanin, especially in the basal cell layer. In the
spreading pigmented AK variant, in addition to the
typical AK histopathologic findings, one may see increased melanin in melanocytes and in the atypical
keratinocytes. Lichenoid AKs are characterized by a
dense, band-like lymphohistiocytic infiltrate at the
dermalepidermal junction with basal cell liquefactive degeneration and occasional Civatte bodies.
In the acantholytic AK variant there are clefts of
lacunae directly above the atypical keratinocytes
with discohesive individual atypical keratinocytes
within the clefts, reminiscent of the changes seen in
Dariers disease. The acantholysis in an acantholytic
AK is a result of the anaplastic changes occurring
in the lower epidermis and loss of normal keratinocytic attachments. The clear cell variant of AK arises
from an excess of cytoplasmic glycogen, which
results histopathologically in marked vacuolation of
the epidermis.
Two different histopathologic reaction patterns may be seen in AKs: namely, epidermolytic
hyperkeratosis (EHK) and the pattern known as
intraepidermal epithelioma of Borst-Jadassohn. In an
otherwise histopathologically classic AK, one may
find incidental changes consistent with EHK. The

Copyright McGraw-Hill Companies, Inc. All rights reserved.

160 Chapter 113: Epithelial Precancerous Lesions

findings of EHK in these AKs most commonly are

focal and isolated and have no clinical importance.
Some authors consider intraepidermal epithelioma
of Borst-Jadassohn to be a specific lesion,23 but
other authors24 believe that this entity actually
represents a type of reaction pattern found in a
variety of lesions, such as AK, pagetoid BD, clonal
seborrheic keratosis, intraepidermal malignant
eccrine poroma, mammary and extramammary
Pagets disease, intraepidermal junctional nevus,
and melanoma in situ. The pattern so named is that
of an intraepidermal neoplasm characterized by
nests and individual cells proliferating or scattered
within normal epidermis. In so-called pagetoid AK,
there are adjacent typical changes of AK, although
differentiating this entity from pagetoid BD is often
quite difficult.
An important point to remember in making the
diagnosis of AK histopathologically is that if the
physician has a strong clinical suspicion of SCC
or basal cell carcinoma (BCC) and the initial cuts
reveal AK, it is prudent to section more deeply into
the block of tissue to ensure that an SCC or BCC
is not missed. In one study additional diagnostic
findings were present on step sections in 23 of 69
specimens (33%) initially diagnosed histopathologically as AK. Of these additional findings, 13% were
indicative of BD, 4% of BCC, and 3% of invasive SCC.
Three variables that correlated with the discovery
of malignancy on step sections were ulceration on
the first level, a clinical diagnosis of skin cancer, and
a history of skin cancer confirmed previously by
biopsy examination.25
Because of the intimate association between AK
and SCC, a number of dermatopathologists have
proposed a more descriptive diagnosis that would
more accurately reflect the nature of this disease.
Cockerell has argued that because this entity is
closely akin to cervical intraepithelial neoplasia
(CIN), terms such as solar keratotic intraepidermal
SCC or keratinocytic intraepidermal neoplasia would
better define AK.1,8 Ackerman26 has argued that
solar keratosis is in fact SCC and should not be distinguished histopathologically. At this time, actinic
keratosis and solar keratosis are the most recognized
terms for this diagnosis, and most authors consider
these lesions precursors of SCC rather than the
malignant entity itself.

Diagnosis and Differential

The diagnosis of AK is accurately made by clinical
examination in most instances when the examiner is familiar with AK and skilled in making such
diagnoses. However, some investigators have commented on the difficulty of accurate and consistent
counting of AK lesions by experts involved in AK
research studies.27 In practice, very few AKs are histopathologically confirmed, and thus few data exist
on the accuracy of clinical diagnosis. The accuracy
of clinical diagnosis is higher in a clinical referral
setting than in the general community setting. One
study involving a referral setting revealed a diagnostic accuracy of 94%, based upon the correct histopathologic diagnosis of 34 of 36 clinically suspect,
typical AKs.28 Another study in a community-based
setting reported a diagnostic accuracy of approximately 80% when random clinically diagnosed AKs
were subject to histopathologic confirmation.29 In
the more recent multicenter Veterans Affairs Topical
Tretinoin Chemoprevention Trial (VATTC), researchers found that the greatest disagreements in the
histopathologic diagnosis of BCC and SCC (keratinocytic carcinomas) between two referral dermatopathologists were in making the diagnoses of AK
and invasive SCC.30
The clinical features of typical erythematous AKs
are not unique to these lesions. The differential diagnosis includes other common lesions as noted in
eBox 113-0.1. Of the clinical AK variants, pigmented
AK is difficult to distinguish from solar lentigo and
sometimes lentigo maligna; spreading pigmented
AK, from lentigo maligna or a large seborrheic keratosis; HAK, from SCC; and cutaneous horn, from the
many entities it may represent.
Clinically, the greatest challenge is to distinguish
an AK from an SCC, because the latter portends
a more serious prognosis and requires different
treatment. There are no reliable clinical criteria for
distinguishing between these two entities, but findings of induration, pain, larger size, marked hyperkeratosis, ulceration, bleeding, rapid growth, and
recurrence or persistence after treatment make the
diagnosis of SCC more likely.
Histopathologically, AK can be confused with
benign lichenoid keratosis (BLK) or lichen planuslike keratosis (LPLK), because both can show vacuolar alteration of the basal cell layer and lichenoid

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

infiltrates. However, closer inspection reveals that in

BLK/LPLK there is no cellular atypia of the keratinocytes as is seen in AK, and often the remnants
of solar lentigo are seen at the periphery of the
lichenoid infiltrate. An atrophic AK can be mistaken
for cutaneous lupus erythematosus, because both
can demonstrate epidermal thinning and basal
cell liquefaction. In cutaneous lupus, however,
other classic features of the disease should also be
present, such as follicular plugging and a patchy
periappendageal infiltrate. Finally, pigmented AK
may be confused histopathologically with lentigo
maligna, but the latter usually has more flattening
of the epidermis, an increased number of single
and nested melanocytes, atypical melanocytes, and
normal keratinocytes.

5-FU therapy has been a mainstay of topical
treatment for AKs since the 1960s (see Chapter
220). 5-FU is a pyrimidine analog that inhibits the
enzyme, thymidylate synthase, thus, interfering in
DNA synthesis and ultimately resulting in apoptosis
and tumor cell death. 5-FU acts selectively to cause
cellular destruction in actinically damaged cells but
not in normal skin. 5-FU treatment of sun-damaged
skin usually causes a brisk inflammatory reaction,
which is felt to contribute to its overall efficacy.
Although many different treatment regimens have
been proposed and used with topical 5-FU, the
historical standard FDA-approved method consists
of twice-daily application of 5-FU cream or solution to the entire affected region, typically for 24
weeks. Spot treatment of specific AK lesions is not
recommended but is a common practice. There is
also some evidence that concurrent treatment or
pretreatment with topical tretinoin is more effective than 5-FU alone for AKs on the extremities.51
Therapy should be continued until the treated area
demonstrates erythema, erosion, crusting, and necrosis, at which time treatment should be discontinued. Different strengths (1%, 2%, 5%) and vehicles
(cream or solution) of topical 5-FU are available.
More recently, a 0.5% micronized 5-FU cream has
become available and FDA approved for treating
AKs. It can be applied once daily for 4 weeks. The
choice of which topical 5-FU formulation to use
depends on physician preference, individual patient
characteristics, and the number and location of AKs
being treated.

Epithelial Precancerous Lesions 161

Anecdotally, topical 5-FU has been thought to

be 75% effective, that is, seven or eight out of ten
treated AKs will resolve with appropriate therapy.52
The long-term remission rates after 5-FU treatment
vary, but the condition frequently recrudesces in a
few years and severely affected patients sometimes
return to baseline.
The once-daily, 0.5% micronized 5-FU cream was
developed in an attempt to make the topical 5-FU
therapy more tolerable. It is FDA approved for the
treatment of multiple AKs on the face and anterior
scalp. Premarketing studies of this medication
showed that more than 75% of treated lesions clinically resolved in 60%80% of patients treated for 4
weeks. One study found an overall 89% reduction
of AKs and a complete clearance rate of 48% in patients treated for 4 weeks with a once-daily application of 0.5% micronized 5-FU cream.53 In a split-face
comparison study of 5% 5-FU and 0.5% 5-FU cream,
similar complete clearance rates of approximately
50% were observed for both therapies.54 The investigators noted a similar reaction rate with erythema,
crusting, and scabbing, but overall the patients
preferred the 0.5% micronized 5-FU cream because
of its once-daily application regimen and a perception of less irritation.
Effective treatment relies greatly on the patients
understanding and compliance with the prescribed
5-FU regimen. The clinician should spend time with
the patient educating the patient about the proper
application of the medication, the duration of
therapy, and the expected outcome. Written handouts reiterating this information and photographs
of the expected reaction are helpful for patients to
view and take home.
Patients should expect discomfort, pruritus,
and burning at the sites of application, as well as
erythema, erosion, crusting, and possibly ulceration beginning at around 1 week into treatment
(eFig. 113-6.1). These and other adverse reactions
to 5-FU are further discussed in Chapter 220. Other
photodamaged areas of the skin that seemingly
have no visible evidence of AKs may also become
quite inflamed during treatment, leading not only
to the treatment of unmasked AKs, but also to the
histopathologically proven rejuvenation of the
skin.55 Much lower cure rates are seen when patient
compliance is poor and the expected end points
are not reached.

Copyright McGraw-Hill Companies, Inc. All rights reserved.

162 Chapter 113: Epithelial Precancerous Lesions

The second agent available for topical field treatment of multiple AKs is imiquimod cream (see
Chapter 220). It works by inducing both the innate
and adaptive arms of the immune system to recognize and destroy AKs. It is available in 5%, 3.75%
and 2.5% formulations. First approved in 1997 by
the FDA for the treatment of genital and perianal
warts, it was approved in 2004 for the treatment
of nonhypertrophic AKs on the face and scalp in
immunocompetent adults. It is also approved for
treatment of superficial BCC (see Chapter 115).
The FDA-approved regimens for the three different strengths of imiquimod are outlined in Box 1132. Treatment with imiquimod cream is associated
with adverse reactions, most notably local skin reactions such as erythema, scabbing, flaking, erosion,
weeping, and vesicle formation. Such reactions are
actually predictive of better ultimate clearance of
the AKs, as the greatest resolution of AKs has been
seen in patients with the most severe reactions.5658
Systemic absorption is minimal, and overall excellent safety margins have been demonstrated with
the topical administration of imiquimod cream.59
Systemic adverse effects, including interferon-like
side effects such as flu-like symptoms, headaches,
and myalgia, have been infrequently reported particularly when large surface areas are treated.
The short-term efficacy of imiquimod cream
for eradicating AKs has been well studied in the
last several years. The first systematic review and
meta-analysis of short-term randomized controlled
trials evaluating its efficacy and safety has been
published.60 For this meta-analysis, five trials57,58,6163
involving a total of 1,293 patients were selected
for inclusion based on their high-quality scores,
minimal risk of bias, validity, adequate treatment
duration, inclusion of patients with multiple AK
lesions, and use of at-home application. Complete
clearance occurred in 50% of patients treated with
imiquimod versus 5% treated with vehicle alone.
Partial clearance (more than 75% of lesions) was
seen in 65% of imiquimod-treated patients versus
11% of vehicle-treated patients.60
Evidence also supports the use of imiquimod
in the treatment of AKs to promote long-term
immune surveillance and lower recurrence rates.
Stockfleth et al56 followed 25 imiquimod-treated patients from a previous short-term study61 for up to
2 more years. Of these 25 patients (one was lost to
follow-up), 8% (n = 2), 16% (n = 4), and 20% (n = 5)

developed new AKs or were lost to follow-up at 12,

18, and 24 months, respectively, after completion of
the initial regimen of imiquimod application three
times a week for 12 weeks. No patients had developed SCC in the imiquimod-treated areas at 2 years
after treatment. Ten of the vehicle-treated patients
from the original study61 were also observed for 1
year after their initial treatment. In this group, 90%
(n = 9) developed new AKs in the vehicle-treated
area, 10% (n = 1) had spontaneous remission of
their AKs, and 10% (n = 1) developed an SCC in the
vehicle-treated area.
In a second study of long-term efficacy, Lee et al64
looked at patients from four previous phase III randomized, vehicle-controlled studies who showed
complete clearance of all of the AKs in the treated
areas at 8-weeks posttreatment. After a median
follow-up of 16 months they found that 25% (n =
19/77) and 43% (n = 23/54) of patients treated with
imiquimod three times per week and two times per
week, respectively, for a duration of 16 weeks experienced a recurrence of one or more AKs in original
treated area.
Imiquimod has been shown to be safe and effective in treating AKs in the organ transplant population, with no evidence of graft rejection and no
trends toward diminished graft functioning.65 Rare
flares of underlying autoimmune disease were
observed during preapproval research studies and
have been documented in the literature.66

The third FDA-approved topical field therapy for
treating AKs is 3% diclofenac gel. Diclofenac is a
nonsteroidal anti-inflammatory drug whose mechanism of action is not completely understood but
is thought to involve inhibition of cyclooxygenase
(COX), a rate-limiting enzyme in the synthesis of
prostaglandins. Diclofenac appears to have greater
affinity for the COX-2 enzyme than for the COX-1
isoform. One theory regarding its role in eradicating AKs relates to the finding that the production of
prostaglandins from arachidonic acid may contribute to UV radiation-induced NMSC. Thus, inhibition
of this cascade by diclofenac may also explain its
efficacy in treating AKs.67
The FDA-approved regimen for diclofenac in the
treatment of AKs is twice-daily application for 90
days. Compared with imiquimod and 5-FU, diclofenac does not cause as much inflammation, irritation, or downtime for the patient. Adverse effects

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

include mild local site reactions, such as erythema,

pruritus, flaking, and possible contact dermatitis. In
a blinded, controlled study involving 96 patients
complete clearance of AKs was observed in 50% of
the diclofenac-treated group and in 20% of the vehicle-treated group 30 days after the conclusion of
therapy.68 At 1-year follow-up, 91% of patients had
at least 75% clearance of the targeted lesions, and
of these patients, 79% showed 100% clearance.68

Future Treatments
The future of treatment for AKs is an area of great
interest and research. Investigators are looking at
reformulating existing topical agents to improve
upon efficacy and to decrease adverse side effects.
Older methods are being challenged in clinical
studies with newer regimens that involve sequential or concurrent therapy with existing AK treatments. Another interesting area of research is looking at the various combinations of lesion-directed
therapies with field treatments, and lastly, new
emerging agents are undergoing rigorous clinical
research to establish their efficacy and safety.8789

Knowledge of the medicinal benefits and poisoning potential of arsenic dates back to ancient
times. Arsenic was introduced into the United
States Pharmacopoeia in 1850, and before its use
was discontinued around 1965, it was employed
medicinally in the United States and Europe in the
form of Fowlers solution, Donovans solution, and
Asiatic pills for treatment of various illnesses such as
psoriasis, asthma, and syphilis. Medicinal exposure
is now basically limited to treatment of tropical diseases, such as African trypanosomiasis, and more
recently to treat various hematologic malignancies.109 Arsenic in opium has been and is still used
for medicinal and recreational purposes in India,
and inorganic arsenic is still found in some traditional Chinese herbal preparations.110 Unfortunately,
these substances can be easily purchased and are
increasingly being used again as homeopathic remedies, resulting in modern day cases of what once
was primarily a disease of historical interest.111,112
Arsenic exposure can occur in a variety of occupations, either by direct exposure to arsenic or
through indirect exposure from contaminated water and landfills. In 1973, it was estimated that more
than 1.5 million workers in the United States were
potentially exposed to arsenic in the workplace.113

Epithelial Precancerous Lesions 163

Occupations that carry risk of exposure include

jobs in the mining, smelting, agricultural, computer
microchip, forestry, electroplating, semiconductor, and glassmaking industries. Environmental
exposure is often obscure and insidious, because
there may be latent periods of up to 50 years before
manifestations of chronic arsenicism appear. Arsenic is routinely found in soil, and it can accumulate
in ground water and well water from these natural
sources. It is becoming a public health concern in
many areas, including Bangladesh, India, Taiwan, Japan, Mexico, Chile, Argentina, Canada, Pakistan, and
the United States.114 Other routes of environmental
exposure include some illegally produced alcoholic
beverages and the burning of pressure-treated
lumber that has been pretreated with chromated
copper arsenate. In all forms of medicinal, occupational, and environmental exposures, longer duration of exposure and higher cumulative dose are
associated with a higher risk for the development
of ArKs.

The association between arsenic exposure and
skin cancer was first noted by Paris in 1822.115 ArKs
were first described on the palms and soles of
individuals in the late 1800s. In 1898, Dubreuilh
categorized ArKs as precancerous lesions.5 The relationship of arsenic ingestion to palmar and plantar
ArKs and to skin cancer has been documented in
several studies that examined wide-scale exposure
to known sources of arsenic.116 In particular, chronic
arsenicism and an increased risk of skin cancer have
been attributed to drinking water that contains
more than 0.6 mg of arsenic per liter.114 The US Environmental Protection Agency has set a maximum
arsenic contaminant level at 50 g/L.116 Ingestion
of as little as 400 mL of Fowlers solution can be
associated with signs of chronic arsenicism.113 A
compilation of the results of 12 reports that linked
malignancy with the ingestion of inorganic arsenic
found that skin cancer occurred in 70% and an
internal malignancy in 6.3% of ingesters.117

Etiology and Pathogenesis

Arsenic is present as either organic or inorganic
compounds as well as in three potential oxidative
states: metalloid, trivalent, and tetravalent. Trivalent arsenicals are the most common and hazardous to humans. The toxicity of these compounds
depends on the accumulation of arsenic in target

Copyright McGraw-Hill Companies, Inc. All rights reserved.

164 Chapter 113: Epithelial Precancerous Lesions

tissues and its metabolism and elimination. Organic

arsenicals are excreted rapidly. Trivalent inorganic
arsenicals are the most acutely and chemically toxic
compounds. Because arsenic is metabolized and
detoxified in the liver via methylation, patients with
preexisting liver disease may be at greater risk for
arsenic-related toxicity.

pink papule with a thick horny layer or crust. When

this crust is removed, the underlying skin appears
erythematous and oozing. Over time, unlike the
relatively stable ArKs, these lesions increase in size,
forming nodular and plaque-like lesions (see eFig.
113-6.2B) that often group together. Approximately
one-third of patients have multiple lesions of BD.110

The mechanisms of arsenic-induced keratoses and

malignancy are not fully understood. Arsenic reacts
with the sulfhydryl groups in certain tissue proteins
and subsequently affects many different enzymes
that are essential to cellular metabolism. Arsenic
has been found to cause chromosomal mutations,
chromosomal breaks, sister chromatid exchanges,
and mutations in p53.118,119 A recent study demonstrated that polymorphisms in a nucleotide excision
repair pathway gene, ERCC2, are associated with an
increased susceptibility to arsenic-induced genotoxic effects and the formation of ArKs.120

Arsenic-induced BCCs are also usually multiple,

and the majority are of the superficial type, although small nodular BCCs are sometimes seen.
These BCCs are found in a random, scattered distribution, primarily on the trunk and in hair-bearing
regions. Clinically, the superficial BCCs are often
indistinguishable from BD.

Clinical Findings
ArKs typically begin as pinpoint papules that are
more easily felt than seen. They develop into small,
20-mm, punctate, yellow, keratotic papules most
commonly seen on the palms and soles in areas of
constant pressure or repeated trauma (eFig. 1136.2A). They preferentially arise on the thenar and
lateral borders of the hands, the sides of the fingers,
and sometimes on the dorsal aspects of the fingers,
overlying the joints. Although it is unusual, ArKs can
be found on more widespread body areas such as
the trunk, extremities, eyelids, and genitalia. ArKs
may also present as slightly elevated, erythematous,
scaly, or pigmented plaques. ArKs on the palms and
soles are more likely to be seen in individuals with
chronic arsenicism caused by medicinal exposure
than in those with arsenicism due to occupational
exposure. Also, individuals with arsenical-related
cancer are more likely to have palmar and plantar
ArKs. The mean latency period for the development
of ArKs varies considerably from 9 to 30 years.121
Other cutaneous neoplasms associated with
chronic arsenicism include BD or SCC in situ, SCC
(see Chapter 114), and BCC (see Chapter 115). Mean
latency periods for the development of BD and
SCC can also be as long as 40 years.121 With arsenic
as the carcinogen rather than UV radiation, these
neoplasms, like ArKs, are distributed in more widespread, random, and nonphotodamaged areas. Arsenical BD often begins as a small flesh-colored to

Arsenic-related SCC can arise de novo or from

malignant transformation of ArKs and BD lesions. In
one study, 55% of SCCs arose from preexisting ArKs
or BD lesions.121 Patients with SCC are more likely
to have been exposed to arsenic earlier in life than
are those without SCC.121 They also are more likely
to have multiple lesions of BD and more numerous
palmar and plantar ArKs. The majority of SCCs have
been found on the distal extremities of the hands
and feet, and it has been hypothesized that irritation and trauma in these areas may increase the risk
of malignant transformation. Clinically, ArKs that
show progression to SCC often present with pain,
bleeding, fissuring, and later ulceration. They tend
to gradually expand in diameter, forming a large
erosion or ulceration, and they can, if neglected,
reach sizes of up to 20 cm.
Other signs of chronic arsenicism include hyperpigmentation primarily affecting the nipples, axillae,
groin, and other pressure points. Within these hyperpigmented patches are often small areas of guttate hypopigmentation, resembling raindrops in
the dust. Diffuse alopecia of the scalp may be present. Longer term, patients may develop blackfoot
disease, which is a peripheral vascular disorder affecting the lower extremities that eventually results
in gangrene. Other systemic findings of chronic
arsenicism include hematologic disorders, liver
disease, sensorimotor neuropathy, hypertension,
diabetes, and other gastrointestinal disturbances. A
variety of internal malignancies have been associated with chronic arsenicism, with long latency
periods of 2050 years. The internal malignancies
linked with chronic arsenicism include lung, urinary
tract, and hepatic tumors, hepatic angiosarcoma,
leukemia, and lymphoma.

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

The histopathologic features of ArKs are essentially the same as those of AKs. No reliable histopathologic criteria can distinguish between the two.
Some cases of ArKs are characterized by marked
vacuolation of the epithelial cells and keratin horn
formation. Also, solar elastosis is usually absent, and
a chronic dermal lymphocytic infiltrate is commonly seen. Likewise, arsenic-related BD, BCC, and SCC
are histopathologically indistinguishable from their
nonarsenic-induced counterparts.
ArKs may be mistaken for other types of punctate
keratoses, such as disseminated punctate keratoderma, Darier disease, corns, and verruca vulgaris.
Disseminated punctate keratoderma usually appears earlier in life. It has also been reported that
upon removal of the keratinous plugs in punctate
keratoderma, small crater-like pits are left behind,
whereas no pits are seen on removal of the keratin
plugs in ArKs. Darier disease presents with characteristic lesions elsewhere. Corns are usually not so
numerous and not so common on the hands. Warts
often demonstrate evidence of thrombosed capillaries upon removal of the surface keratin.
A diagnosis of ArKs and chronic arsenicism should
be considered when numerous characteristic
keratoses are seen on the palms and soles or when
multiple lesions of BD, SCC, or BCC are found on
an individual, especially when these lesions are
in nonsun-exposed regions of the body. In most
patients with such neoplasms, palmar and plantar
keratoses will also be present. Such patients should
be questioned about previous occupations, living
conditions, and environmental and medicinal exposures to elicit a history of potential arsenic exposure
anywhere from 10 years to 40 years previously.
Biopsy of any changing ArK, erythematous nodule,
plaque, or ulcer on the body should be performed
to ensure that BD, BCC, or progression to SCC is
not present. In addition, laboratory tests to order
include a complete blood count, renal and liver
function tests, hair and nail analysis to measure arsenic concentration, and nerve conduction studies
if a sensorimotor neuropathy is suspected.

Etiology and Pathogenesis

Thermal keratoses (TKs) are keratotic lesions
produced in the skin by exposure to infrared radiation. They have the potential to develop into SCC.
Various sources of heat are implicated in causing

Epithelial Precancerous Lesions 165

these lesions, and several classic syndromes have

been reported. Most notable are the kangri basket
cancers in Kashmir, the kang cancers in China, the
railway cancers in Britain, and the peat fire cancers
in Ireland. The kangri is a pot that contains some
type of fuel, such as charcoal, which is held against
the thigh or lower abdomen as a source of warmth.
In Kashmir, the malignancies would arise in the skin
over the lower abdomen and thighs, where the baskets were in contact with the body. Kang are brick
beds used primarily in the northeastern part of
China that are usually heated by pipes connected
to coal-burning stoves. The skin overlying the greater trochanter is where the cancers were most likely
to develop. Heat was reported to be responsible
for cancer of the skin overlying the shins of railway
engine drivers in England. In Ireland, more than 150
women were reported to have lesions of erythema
ab igne and SCC on the lower legs. These women
all reported many years of sitting close to an open
fire of burning peat to stay warm. Because peat is a
less efficient fuel than coal, these women had to sit
closer to the source to attain adequate warmth. The
men did not experience the same cutaneous effects
because they primarily wore thick, long trousers. In
the United States, TKs, BD, and the development of
SCCs have been noted in persons after long-term
exposure to coal- and wood-burning stoves used
as sources for heat and cooking.126 The incidence
and prevalence of TKs are not known. Likewise, the
pathogenesis of these lesions is not fully understood, and the percentage of TKs that will progress
to invasive SCC is also unknown. Erythema ab igne,
the precursor lesion, has been reported as a result
of long-term exposure to laptop computers and
electric heating pads.127,128

Clinical Findings
Prolonged exposure to infrared radiation and to
sources of heat can clinically produce the characteristic lesion of erythema ab igne, which manifests as an erythematous to brownish, fixed, thick,
reticulated patch on the skin overlying the area
that has been habitually exposed to heat (see Fig.
75-21). It differs from livedo reticularis in that it is
more brownish, usually nonblanchable, and fixed.
The sites of predilection are the back and abdomen
from exposure to hot water bottles and heating
pads or blankets, and the lower legs from exposure
to fires or heating units as sources of warmth and
most recently the anterior thighs from chronic
exposure to laptop computers.128 After many years
of such exposure, TKs may appear within these

Copyright McGraw-Hill Companies, Inc. All rights reserved.

166 Chapter 113: Epithelial Precancerous Lesions

patches of erythema ab igne. SCC in situ within

such patches has also been reported, and as with
solar-induced AKs, progression to invasive SCC from
TKs can occur. Clinically, TKs present as hyperkeratotic papules and plaques.

The histopathologic features of TKs are similar to
those of AKs, including alternating hyperkeratosis
and parakeratosis, keratinocyte atypia, sparing of
the appendageal units, and, interestingly, even
dermal elastosis and vascular ectasia. In one study,
biopsy specimens from erythema ab igne patches
without apparent TKs showed evidence of keratinocyte atypia.129 Merkel cell carcinoma has also been
reported to arise alongside SCC within a patch of
erythema ab igne.130

Diagnosis and Differential

Biopsy should be performed on hyperkeratotic
papules or plaques that develop within a longstanding patch of erythema ab igne to confirm the
diagnosis and to rule out progression to SCC.

Prognosis and Treatment

There is a paucity of information in the literature
regarding the risk of progression of TK to invasive
SCC and the general prognosis of thermal SCC. Similarly, little has been written about the treatment of
these lesions. Surgical excision, curettage with or
without electrosurgery, cryosurgery, and possibly
laser therapy appear to be reasonable therapeutic
options. Patients with erythema ab igne should be
counseled about the need to discontinue exposure
to whatever heat source they have been in contact
with, and they should be followed long term with
physical examinations to identify early changes
suggestive of TK or SCC.

Hydrocarbon Keratoses
Etiology and Pathogenesis
Hydrocarbon keratoses (HKs), also known as tar
keratoses, pitch keratoses, and tar warts, are precancerous keratotic skin lesions that occur primarily
in persons who are exposed occupationally to
polycyclic aromatic hydrocarbons (PAHs). SCC and

keratoacanthoma have also been linked to such

exposures. PAHs are produced by the incomplete
combustion and distillation of coal, natural gas, and
oil shale. They are found in tar, fuel oils, lubricating
oils, oil shale, and bitumen.115 Patients at risk for
exposure to such compounds include workers in
tar distillation and shale extraction, and workers
exposed to creosotes, which are one of the main
distillate fractions of coal tars and coal tar pitches.131
Creosotes are used extensively as wood preservatives. Roofers, asphalt workers, road pavers, and
highway maintenance workers are exposed to PAHs
from fumes formed during the heating of bitumen.131 Other groups of workers at risk for exposure
to coal tar products include railway workers, brick
masons, and diesel engineers.115 Finally, persons exposed to mineral oils, cutting oils, and coolants also
are at risk for the development of HKs and SCC.131
The most infamous occupational exposure to
PAHs was that of chimney sweeps, who developed
scrotal SCC or Potts tumor (named after Dr. Percivall Pott, a London surgeon who described the
condition in 1775).132 Mule spinners (a mule spinner is a yarn-producing machine invented in 1779
by Samuel Crompton) have also been reported to
develop SCC of the scrotal skin.115 Mule spinners
frequently have their thighs and scrotum in contact
with an oil-covered bar on the mule or yarn producing machine.
HKs, SCCs, and keratoacanthomas develop in
persons whose skin is exposed to PAHs over long
periods of time. The risk of developing these precancerous and cancerous lesions is associated with
dermal exposure to PAHs. Inhalation of PAHs is not
associated with skin cancers, although lung cancers
have been reported with such exposures. PAHs are
known carcinogens that damage DNA and potentially cause mutations in the p53 gene, at least in
lung cancers.132 UV radiation exposure has been
considered a cofactor in the development of HKs
and related skin cancers, but one study found that
the majority of HKs and SCCs in a large group of
tar refinery workers developed in nonsun-exposed
regions of the body, which puts this belief into
question.133 The duration of exposure before the
development of HKs ranges from 2.5 to 45 years.134

Clinical Findings
HKs present as small, oval-to-round, gray, flat
papules that are easily removed without bleeding. These papules can then become larger and

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

more verrucous in appearance, and eventually may

develop into invasive SCC. Sites of predilection include the face, nostril rims, upper lip, forearms, volar
wrists, dorsal feet, lower legs, vulva, and scrotum.133
Of note, the lesions on the nostril rims, upper
lip, genitalia, and forearms are not in the typical
sun-exposed sites of the body. Thus, keratoses or
malignancies in these regions should prompt the
physician to inquire about potential occupational
hazardous exposures. Other cutaneous findings
related to PAH skin exposure include patchy hyperpigmentation, acne, and telangiectases at sites
of exposure. The presence of acneiform lesions on
the forearms and thighs should alert one to the
possibility of PAH exposure. Facial features associated with long-term tar exposure include periocular
fibromas, brownish or slate-gray discoloration,
comedones, and follicular plugs.133 PAHs have also
been reported to cause internal malignancies, such
as lung and bladder cancers.131

Histopathologically, HKs appear similar to AKs
and ArKs, as previously described. Often, HKs show
more of a progression to full-thickness atypia or a
bowenoid appearance. In early HKs, it has been said
that distinction between benign and malignant
epidermal changes can be difficult.

Diagnosis and Differential

HKs, with their grayish keratotic appearance, may
be mistaken for verruca vulgaris, stucco keratoses,
or even ArKs. A high index of suspicion is needed
to make the diagnosis of HKs, because histopathologically they are difficult to distinguish from AKs
and ArKs. The distribution of the lesions and other
cutaneous findings of tar-exposed skin can aid in
the diagnosis and lead to targeted questioning of
the patient.

Prognosis and Treatment

The exact risk that an HK will progress to SCC is
unknown. Most recommendations in the literature call for biopsy and surgical removal of HKs,
especially in the vulvar and scrotal regions and on
mucosal surfaces, where the risk of early metastasis
from SCC is greater. Cryosurgery and curettage with

Epithelial Precancerous Lesions 167

or without electrosurgery are probably reasonable treatment options for HKs at other cutaneous
sites. Prevention is the key for workers exposed to
PAHs in their occupations. Long-term follow-up of
individuals with previous known PAH exposures
is essential for early diagnosis and treatment of lesions and for early detection of internal malignancy
in those at risk.

Chronic radiation keratoses (CRKs) are precancerous
keratotic lesions that may arise on the skin many
years after exposure to ionizing radiation. They
have been described in persons exposed to medical
ionizing radiation, including grenz ray therapy for
benign skin disorders, such as acne vulgaris and
tinea capitis; in medical personnel, dentists, and
physicians who have X-rays over many years115; in
other professionals who handle radioactive materials in their workplaces; on the fingers of persons
who unknowingly wore gold rings contaminated
by radioactivity (eFig. 113-6.3)135; in survivors of
the atomic bomb in Japan136; and in the victims
of nuclear accidents such as the one in Chernobyl,
Ukraine.137 For the purpose of the following discussion, radiation refers to ionizing radiation; thus, the
UV and infrared spectra of radiation are excluded.

Clinical Findings
The primary neoplasms resulting in the skin from
exposure to ionizing radiation are CRKs, BCCs, and
SCCs. Cutaneous melanomas and sarcomas rarely
occur. CRKs and BCCs are seen more commonly
than SCCs, but any given CRK can potentially evolve
into invasive SCC. The site and penetration of the
ionizing radiation determine which type of tumor
develops; palms, soles, and mucosal surfaces are
said to be favored locations for CRK. The neoplasms
that develop are often multiple, heal poorly, and recur frequently.115 CRKs usually develop within areas
of chronic radiation dermatitis or in persistent ulcerations arising in such chronically damaged areas of
the skin, but they can also develop on previously
irradiated areas of the skin that clinically appear
normal. They appear as hyperkeratotic papules
or plaques (see eFig. 113-6.3). The latency period
from the time of exposure to the development of
CRKs and radiation-induced skin cancers is typically
long, having been reported to range up to 56 years.

Copyright McGraw-Hill Companies, Inc. All rights reserved.

168 Chapter 113: Epithelial Precancerous Lesions

The latency period for the development of these

tumors is inversely proportional to the radiation
dose. Patients exposed to ionizing radiation are also
at risk for the development of internal malignancies,
such as those of the breast, bone, soft tissue, uterus,
larynx, pharynx, thyroid, and maxillary sinus.115

Histopathologically, CRKs display keratinocyte
dyskeratosis with hyperchromatic nuclei and abnormal mitoses. The dermis often demonstrates findings consistent with long-term radiation-induced
changes, such as hyalinization of the collagen
bundles, thickening and occlusion of deep dermal
blood vessels, endothelial cell atypia, and destruction of the pilosebaceous units.

Prognosis and Treatment

To date, no clinical trials have compared different
lesion-targeted therapies on radiation port sites.
Surgical excision and Mohs micrographic surgery
are often the treatments of choice, because radiation-induced SCCs have a higher potential for metastases. Recent interest has arisen in considering
field treatment of the entire radiation port site with
agents such as 5% imiquimod cream, diclofenac gel,
topical 5-fluorouracil (5-FU), or PDT, but no studies
investigating these agents for such use have been
reported. There is also concern that such field treatments could induce a radiation recall reaction.138
Persons with ionizing radiation exposure should be
followed with skin examinations at regular intervals
for the rest of their lives, because the latency period
for the development of cutaneous neoplasms is
quite long, and the number of skin cancers in such
patients tends to increase with time after irradiation. Evaluation for the development of internal malignancies, for which this exposure carries high risk,
should also be considered in the long-term followup of these patients. Finally, these patients should
be counseled to avoid other potential carcinogens,
such as UV radiation.


Chronic scar keratoses (CSKs), also known as chronic
cicatrix keratoses, are precancerous keratotic lesions
that arise in long-standing scars. CSKs can arise in
association with a number of different chronic scar-

ring processes, such as chronic ulcers, burn scars,

pilonidal sinuses, draining sinuses, chronic osteomyelitis, vaccination scars, chronic hidradenitis
suppurativa, old frostbite scars, and scarring acne

In 1828, Jean-Nicholas Marjolin described chronic
ulcers that arose from scar tissue, without referring
to them as malignant.139 Today, the term Marjolins
ulcer is synonymous with malignant transformation of chronic ulcers, sinus tracts, and burn scars,
although the term is used most frequently to
describe malignant changes within burn scars. It
is estimated that 2% of burn scars will undergo
malignant degeneration. Most burn scar carcinomas are SCCs, but cases of BCC, melanoma, sarcoma,
and malignant fibrous histiocytoma have also been
reported.140 Two types of Marjolins ulcer have been
described: an acute form and a chronic form. The
acute type develops within 1 year from the time of
injury, whereas the chronic form develops longer
than 1 year after the injury.141 The average latency
period from injury to onset of malignancy for
the chronic form is 36 years, with a range of 175
years.142 The latency period is inversely proportional
to the patients age at the time of the burn injury.

The pathogenesis of a premalignant keratosis or
a carcinoma arising in a chronic scar is not known.
Proposed mechanisms for malignant degeneration within a burn scar include production of a
carcinogenic toxin with the burn injury; immunologic privilege within scarred tissue that allows
unchecked tumor growth; chronic irritation leading
to malignancy; multistep carcinogenesis involving initiation, promotion, and progression; and
burn-induced DNA damage leading to malignant
transformation. Characteristics of the burn scar that
predispose to malignant transformation include a
prolonged healing phase, repetitive trauma, and
graft rejection.

Clinical Findings
CSKs present as hyperkeratotic papules, plaques,
or erosions within the scarred skin (eFig. 113-6.4). In
terms of burn scars, two clinical types of carcinoma

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

have been described. The more common presentation is of a flat, ulcerative, indurated lesion with
elevated borders and surrounding induration. The
less common type is an exophytic, papillomatous
lesion resembling granulation tissue.141 Bleeding,
pain, and a foul odor may also be present. Typically,
chronic and recurrent ulcerations have been present in the burn scar before malignant degeneration,
and the malignancy often develops at an ulcer margin. One important fact about burn scar carcinomas
is that usually only a portion of the ulcer undergoes
malignant transformation. The remaining ulcer
persists as a nonhealing wound, and thus the rate
of false-negative results from biopsies of the ulcer
is quite high.140 Sites of predilection for burn scar
carcinomas are the extremities and areas overlying
joints, presumably because of repeated trauma in
these areas.

Histopathologically, most scar keratoses range
in pattern from several atypical keratinocytes to
full-thickness epidermal atypia or in situ SCC. The
majority of burn scar carcinomas are of the well-differentiated squamous cell type. When the diagnosis
is based on histopathologic features, scar keratoses
may be confused with pseudoepitheliomatous
hyperplasia, a benign reactionary process.

Diagnosis and Differential

Biopsy should be performed on any persistent lesion, erosion, or ulceration within a scar. In addition,
for burn scar carcinomas, because only a portion
of the ulcer may be malignant, multiple biopsies
or excisional biopsies are recommended if the
suspicion of carcinoma is high, so the diagnosis will
not be missed.

Prognosis and Clinical Course

CSKs may show a gradual progression to in situ
carcinoma and frank SCC. Prompt diagnosis is
important, because once invasive SCC develops in a
scar, it can quickly become aggressive, with a metastatic rate estimated at approximately 35%, much
higher than for actinically induced SCC.141 Lesions
occurring on the lower extremities have the highest rates of metastases, averaging 50%,143 whereas

Epithelial Precancerous Lesions 169

those on the head, neck, and upper extremities

fare better. Lymph node metastasis on presentation is the single most important indicator of poor

Treatment and Preventiion

Burn scar carcinoma may be prevented by meticulous burn wound care, early skin grafting, prompt
treatment of infections, avoidance of contractures,
and early excision of any degenerative changes
within the scarred area, with step sections through
the specimen to ensure that a malignancy has not
been missed. Excision with wide margins (at least 2
cm) is the treatment of choice for burn scar carcinoma.141,143 Radiation therapy and topical chemotherapy have not been beneficial in the treatment
of these lesions.141,143

Reactional keratoses (RKs) are premalignant
keratotic lesions that may arise in a variety of longstanding inflammatory processes such as cutaneous lupus erythematosus, necrobiosis lipoidica,
porokeratosis, erythema elevatum diutinum, granuloma inguinale, epidermolysis bullosa variants,
pemphigus vulgaris, lichen sclerosus, lichen planus,
and chronic deep fungal infections. Clinically, RKs
present as hyperkeratotic papules or plaques. As
with other precancerous keratotic lesions, a continuum from RK to SCC in situ to invasive SCC may
be seen. Once invasive SCC has appeared, there is a
significant risk that metastatic disease will develop,
similar to that with scar keratoses. The main diagnosis to consider in the clinical and histologic differential is pseudoepitheliomatous hyperplasia.

Psoralen plus ultraviolet A (PUVA) keratoses are
precancerous keratotic lesions that arise in individuals exposed to PUVA for the treatment of a
variety of cutaneous conditions. UVA light alone is
thought to play a role in skin cancer formation, but
the combination of UVA with psoralen is associated
with a dose-dependent increased risk of premalignant keratoses and other cutaneous malignancies,
such as SCC, melanoma, and, possibly, Merkel cell
carcinoma (see Chapters 112 and 238).

Copyright McGraw-Hill Companies, Inc. All rights reserved.

170 Chapter 113: Epithelial Precancerous Lesions

The term intraepithelial neoplasia (IN) is defined as
a precancerous squamous epithelial lesion of the
lower anogenital tract. The terminology of these
precancerous lesions has been confusing, with various names such as squamous intraepithelial lesion
(SIL), erythroplasia of Queyrat (EQ), genital Bowen
disease (GBD), bowenoid papulosis (BP), and SCC in
situ being used in the past. It is now recommended
that the following terms be used, as outlined
in eBox 113-5.1. All of these INs have a disease
spectrum ranging from mild-dysplasia to severedysplasia, but with no evidence of invasion into the
dermis. For the purposes of this chapter, only vulvar
intraepithelial neoplasia (VIN), anal intraepithelial
neoplasia (AIN) and perianal intraepithelial neoplasia (PaIN), and penile intraepithelial neoplasia (PIN)
will be discussed. Because most dermatologists
consider BP to be a distinct clinical entity, with a
classic presentation in mostly younger individuals,
a better prognosis, and a greater chance for spontaneous regression, this entity has been discussed
separately under Section Viral-Associated Precancerous Lesions.
VIN is a precursor lesion of vulvar SCC. The original
classification of VIN, dating back to 1986, identified
three grades of VIN (VIN1, VIN2, VIN3), based upon
the degree of atypical cells within the vulvar squamous epithelium. This was analogous to the CIN
spectrum. In 2004, the vulvar oncology subcommittee of the International Society for the Study of
Vulvar Diseases (ISSVD) proposed a new classification system for VIN, designating the term VIN be
used only for high-grade lesions (formerly VIN2 and
VIN3) that have the potential to progress to vulvar
SCC. Thus, the older designation of VIN1 has been
eliminated, primarily because such VIN1 lesions are
thought to represent benign, reactive or HPV-related effects, and not precancerous lesions.171

The incidence of VIN is increasing globally. It is primarily a disease of younger women, who comprise
about 75% of all cases.172 The incidence of vulvar
cancer is also increasing, but at a slower rate.

Etiology and Pathogenesis

VIN is classified into three main groups: (1) VIN,
usual type (warty, basaloid, or mixed), (2) VIN, differentiated type and (3) VIN, unclassified type (eBox
113-5.2). The VIN, usual type incorporates what was
previously called VIN2, VIN3, or SCC in situ. These
lesions are associated with hrHPV16,18,31 infection
and are further histopathologically subcategorized
into warty, basaloid, or mixed subtypes. VIN, usual
type primarily occurs in younger, premenopausal
females, and additional risk factors are tobacco use,
other sexually transmitted diseases (STDs) and immunosuppression. These lesions are both multifocal
and multicentric in their presentation, with more
extensive disease displaying confluence of lesions.
VIN, differentiated type is much less common and
accounts for only 2%10% of all previously designated VIN3 cases.173 This lesion primarily affects
postmenopausal women, is usually unifocal and
unicentric, and is often associated with lichen sclerosus. Infection with HPV is uncommon and it is not
thought to play a role in its pathogenesis.

Clinical Findings
On examination, VIN, usual type can present with
a variety of clinical appearances, such as an erythematous, well-defined plaque; a verrucous white
plaque; grouped pigmented papules or plaques;
erosions or ulcers. These lesions are most often
multifocal and may demonstrate extensive involvement of the perineum and the adjacent skin. Over
50% of patients with VIN, usual type, have associated CIN. Individual lesions of VIN, differentiated
type may present as an erythematous or hyperkeratotic plaque; a verrucous papule or plaque; or as a
nodule or ulcer.
Approximately half of all patients with VIN are asymptomatic and the abnormal lesion is discovered
on a routine gynecological examination. Symptomatic patients typically present with pruritus as their
main complaint. Other symptoms may include pain,
burning, dysuria, bleeding, discharge, a palpable
lesion, or a persistent ulcer. Any suspicious lesion
should be biopsied. A punch or excisional biopsy
is preferred, in order to determine the depth of
the lesion and to not miss invasive SCC. A patient
who has multiple or multifocal lesions will require
multiple biopsies. Histopathologically, VIN is charac-

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

terized by disorderly maturation of keratinocytes

within the epithelial layer, nuclear atypia, coarse
nuclear chromatin, and multiple mitotic figures
above the basal layer.

Differential Diagnosis
The clinical differential diagnosis includes other
inflammatory disorders, such as lichen sclerosus,
lichen planus, candidiasis, condyloma acuminata,
and perhaps melanoma in those cases of pigmented lesion presentation.

Prognosis and Clinical Course

If left untreated, VIN may persist, regress or progress to vulvar SCC. In a literature review of ten studies that involved 61 untreated VIN patients and 27
with macroscopic residual disease, 9% were found
to have progressed to invasive vulvar SCC over the
course of 18 years.174 Spontaneous regression has
been described primarily in young women (mean
age of 20 years). Despite treatment of VIN it may
recur in one-third of women and 4%8% may go on
to develop locally invasive SCC. Long-term surveillance of these women is crucial.

The main goals of treatment are to prevent the
progression to invasive SCC; to relieve patient
symptoms; and to retain as much of the vulvar
anatomy and function as possible. Unfortunately,
there are few, sufficiently powered, long-term,
randomized trials looking at the management of
VIN. Available treatment options include surgical
excision, partial or total vulvectomy, laser ablation,
and field treatments such as topical 5% imiquimod
and 5-FU creams. Treatment with 5% imiquimod
has been shown to be effective in treating VIN and
is preferred by many experts over topical 5-FU, if
topical therapy is chosen.175179 Obviously, any treatment discussions would be handled in conjunction
with the patients gynecologist. Lastly, any patient
in whom VIN is diagnosed will require a complete
gynecologic exam because of the high risk of associated CIN.
The treatment of VIN with currently available HPV
vaccines is an exciting, ongoing area of research.180
A recent study, using an HPV 16 vaccine in females
with the usual type of VIN, showed that at 1 year
15/19 patients had a partial (n = 6) or complete (n =

Epithelial Precancerous Lesions 171

9) clearance of their VIN lesions, and that at 2 years

these 9 complete responders had a sustained complete response.181 This study is the first of its kind to
show that effective immune responses can be generated against precursor lesions, and in conjunction
with the imiquimod treatment data,175177 it also
suggests a stronger role for such immunotherapy in
the treatment of VIN.

Prevention of VIN and progression to vulvar SCC is
now a reality. It is believed that approximately twothirds of VIN could be prevented in younger women
with the prophylactic use of the currently available
HPV vaccines180,182,183 (see Chapter 196).

Anal (AIN) and Perianal (PaIN)

Intraepighelial Neoplasia
AIN is an HPV-associated precursor lesion of anal
SCC. PaIN is an SCC in situ cutaneous lesion of the
perianal skin. AIN has also been referred to as anal
squamous intraepithelial lesion (ASIL), anal dysplasia,
and in precancerous lesions of the perianal skin
(PaIN) the terms GBD, SCC in situ, and BP have all
been used interchangeably. AIN is classified in the
same nomenclature as that used to describe CIN.
AIN1, AIN2, and AIN3 correspond to histopathologic
mild, moderate, and severe dysplasia, respectively.
AIN1 is not considered a direct precancerous lesion
of anal SCC, but it can progress to AIN 2 and AIN 3,
both of which are considered precancerous lesions.

Epidemiology, Etiology, and

AIN and anal SCC are increasing in the general
population and are most prevalent in HIV-positive
men who have sex with men (MSM), and also in
HIV-positive females. The greatest risk factors for
developing AIN, in both males and females, are
infection with HPV, receptive anal intercourse, HIV
infection, and decreased CD4 counts, although
restoration of CD4 counts to normal with highly
active antiretroviral treatment (HAART) does not
seem to cause regression of AIN in HIV-positive
persons. Other risk factors include immunosuppression, tobacco use, and abnormal cervical cytology
in females. The high-risk subtypes of HPV that are
associated with AIN and anal cancer are HPV 16, 18,
31, and, less so, 33.

Copyright McGraw-Hill Companies, Inc. All rights reserved.

172 Chapter 113: Epithelial Precancerous Lesions

Clinical Findings
Patients with AIN may be asymptomatic or may
present with localized symptoms, such as pruritus,
pain, bleeding, tenesmus, and discharge. On physical exam, AIN can present in a variety of ways. The
dermatologists role is to inspect the perianal skin
for suspicious lesions, such as well-demarcated erythematous pink plaques, grouped hyperpigmented
papules or plaques, verrucous papules or plaques,
and persistent erosions or ulcers. Findings of such
suspicious lesions should prompt an excisional or
punch biopsy, as well as referral to a specialist in
this field of medicine for complete anal exam with
high-resolution anoscopy and cytologic exam. Females should also undergo a complete gynecologic
exam to rule out CIN.
No formal AIN screening guidelines have been
formulated or accepted, but regular screening for
AIN is recommended for all HIV-positive individuals;
all MSM; females with a history of high-grade cervical, vulvar, or vaginal IN; and perhaps immunosuppressed patients, such as transplant recipients. Such
patients should be screened by physicians who are
experienced in the diagnosis and management of
AIN. Dermatologists should identify such high-risk
patients and perform perianal examinations periodically.

The histopathology of PaIN is basically that of SCC
in situ. AIN and CIN share similar histopathologic
features, as described using the 2001 Bethesda
classification.184 High-grade AIN2 and -3 are evident
when the abnormal epithelial basaloid cells replace
more than half of the entire epithelium.

Prognosis and Clinical Course

Relatively little is known about the natural history
of high-grade AIN, other than it rarely spontaneously regresses, regardless of HIV status. Low-grade
AIN1 may spontaneously regress or progress to
high-grade AIN2 or AIN3. Risk factors for progression in MSM include HIV infection, low CD4 counts,
anal HPV infection, and high levels of oncogenic
HPV subtypes. Also, despite treatment, high rates of
recurrence are the norm and thus ongoing surveillance is mandatory.

The treatment goal in AIN is to prevent progression to anal SCC, but there is a lack of good evidence that screening for and treating high-grade
AIN can prevent this progression. There is also
limited data on the efficacy of treatments for AIN.
Many experts recommend that individuals with
high-grade AIN receive treatment, but again, no
definitive algorithms exist. True, high-grade AIN is
best managed by an expert in this field of medicine.
Treatment options for PaIN include similar treatments as described for other cutaneous SCC in situ
A recent literature review to assess the efficacy of
5% imiquimod in treating AIN found that five cohort and case reports have been published to date,
all limited to HIV-positive MSM.184 A pooled analysis
of these studies demonstrated a complete response
of 48%, with about a 36% recurrence rate.185

Prevention of AIN is an exciting possibility with
the advent of HPV vaccines, but this is still undergoing investigation. Clinical studies are also underway
to see if HPV vaccines can induce regression of
high-grade AIN.186
PIN is a precancerous lesion that may progress
to invasive SCC of the penis. Older, synonymous
terms for this in situ carcinoma of the penis are BD,
GBD, EQ, and BP, but many experts in this area now
recommend that PIN be the preferred name for this

The exact incidence and prevalence of PIN are not
known. It is thought to be uncommon in the Western world, but it has been increasing significantly
in some developing countries. Of all the penile
cancers, in situ SCC and invasive SCC are the most
common entities, with SCC representing the majority of invasive penile cancers (~95%).148,187

Etiology and Pathogenesis

PIN is primarily a disease of older, uncircumcised
males, peaking in the sixth and seventh decades
of life. Penile SCC is rare in males circumcised at or
around the time of birth. In a recent study of 213

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

adult males with penile SCC, only two (2.3%) with

invasive SCC and sixteen (15.7%) with PIN were
circumcised as newborns.148 Risk factors for developing PIN and penile SCC in uncircumcised males
include poor hygiene, smegma retention, phimosis,
and other chronic inflammatory and infectious
conditions affecting the penis. Other risk factors are
coinfection with certain oncogenic HPV subtypes,
primarily HPV-16, but also HPV-18, 31, and 33;
penile lichen sclerosus (LS), also known as balanitis
xerotica obliterans (BXO); HIV infection; immunosuppression; tobacco use; and ultraviolet (UV) radiation exposure (PUVA and recreational).187
It has been hypothesized that there are two different and independent pathways in the development
of PIN and penile SCC: an HPV-positive pathway
and an HPV-negative pathway. Approximately
40%45% of all penile cancers are thought to have
an HPV association. The HPV-positive pathway is
thought to result from probable sexual transmission of and acquired infection with oncogenic HPV
subtypes in young adults. This pathway morphologically is believed to result in a basaloid/warty
subtype of penile SCC that histopathologically
displays small basaloid cells with poorly defined
borders or koilocytosis. These warty/basaloid types
of penile SCC display the strongest association with
oncogenic HPV infection (70%100%), and HPV-16
is the predominant subtype.148
The HPV-negative pathway is mostly unrelated
to HPV infection and primarily affects older males.
Morphologically, this pathway results in the usual
type or differentiated squamous/keratinized penile
SCC that histopathologically shows a keratinizing
epithelium with variable degrees of cellular atypia.
This usual type of penile SCC comprises the majority (49%) of all penile SCCs. Approximately 30% of
such non-HPV-associated penile SCC cases are preceded by LS, but in some cases there is coexistence
of LS and hrHPV infection.148

Clinical Findings
PIN and penile SCCs most often originate on the
mucosal surfaces of the penis, such as the glans,
coronal sulcus, prepuce, urethral meatus, and the
mucosal surface of the foreskin. Less than 5% of
PIN and such cancers arise on the shaft of the penis.
The clinical presentation of PIN is variable and depends upon the anatomic site of involvement. The
mucosal presentation has historically been called
EQ and the shaft presentation has been called GBD.
PIN presenting on the mucosal sites often presents

Epithelial Precancerous Lesions 173

as a well-demarcated, glistening, erythematous,

velvety plaque or plaques, as originally described
by Queyrat in 1911 (eFig. 113-8.1).188 Patients with
this EQ variant of PIN may present with symptoms
of pain, pruritus, bleeding, crusting and difficulty in
foreskin retraction. The EQ PIN variant is associated
with HPV 8 in almost all cases and concomitant HPV
16 in most.189 This EQ variant of PIN more often (up
to 30%) progresses to invasive SCC compared with
the GBD variant (3%6%),189 but it usually progresses slowly and the presence of ulceration may herald
the progression to invasive SCC.
The GBD variant of PIN typically presents on the
shaft of the penis as a single, erythematous or variably pigmented, well-demarcated plaque. It may
also display secondary scaling, oozing, crusting, or
erosion. The development of an ulcer, verrucous or
nodular growth, may herald its progression to invasive SCC. Any suspicious or persistent lesion of the
penis should be biopsied for definitive histopathologic diagnosis. Other clinical entities to consider in
the differential diagnosis of PIN are listed in eBox

Histopathologic examination of both the EQ and
GBD variants of PIN are mostly similar and that
of SCC in situ. However, in the EQ variant there is
often epithelial hypoplasia, fewer multinucleated
and dyskeratotic cells, and many more plasma cells
in the dermal infiltrate. Histopathologic studies of
penile SCC cases have observed a sequence from
low-grade PIN to high-grade PIN in many (62%)

Prognosis and Clinical Course

GBD and EQ are histopathologically identified as
high-grade PIN and may progress to penile SCC.
Progression of PIN to invasive SCC is likely in untreated lesions and more common in the EQ variant
versus the GBD variant, although this is not always
the case. Treatment of PIN is warranted to prevent
progression to invasive penile SCC. Once invasive
SCC has occurred, approximately 20% of patients
will display evidence of regional lymph node
involvement or more distant metastases.191 Dermatologists can play a role in the early diagnosis and
management of disease.

Copyright McGraw-Hill Companies, Inc. All rights reserved.

174 Chapter 113: Epithelial Precancerous Lesions

Therapeutic measures for treatment include wide
local excision, Mohs micrographic surgery, laser ablation, and topical 5-FU and 5% imiquimod cream
though evidence for use of the two latter topical
therapies is limited to small series and case studies.
Each case must be individualized and based upon
patient characteristics and preferences, PIN type
and size, physicians skills set, and optimal treatment with the most preservation of anatomy and
function. Life-long surveillance of patients with all
forms of PIN is essential to ensure early detection of
recurrence. Patients should also be advised to initiate partner screening for CIN, VIN, and AIN.


malignant potential,208 but conflicting findings have

more recently been published regarding this correlation.195,209,210
Examination of the oral mucosal surface should
ideally be part of a patients total body skin examination. Such an examination is more important
in patients with a history of tobacco use or combined tobacco and alcohol use, or a prior history
of documented OL, OE, or OSCC. In patients with
definitive OL or possible OL, an examination of the
lymph nodes should also be performed. In addition, patients with definitive OL should be followed
closely, at regular intervals, for the rest of their lives,
because they are at risk for malignant transformation of the lesion and for new lesions.

Additional diagnostic tools

Several preventive strategies for penile SCC have

been proposed. Neonatal circumcision has been
well established as an effective prophylactic measure for penile cancer, but societal, cultural, and religious beliefs have often hampered the widespread
implementation of this procedure. More studies are
needed to determine if circumcision in older males
can play a role in primary or secondary prevention.
In addition to neonatal circumcision, four studies,
from four different countries, have shown a significant decrease in penile cancer without implementing circumcision, mainly by advocating better
education and by improving sanitation.192 Other
preventive measures that are advocated include
good hygiene, stopping smoking, preventing phimosis, treating chronic inflammatory and infectious
conditions, and avoiding genital exposure to PUVA
or UVR. Lastly, HPV vaccination in young males is
being investigated as a prophylactic measure, but
to date no efficacy studies have been published.

Physical examination
OL can occur as an isolated single lesion, as multiple separate lesions, or as diffusely outlined lesions.
It is recommended that OL lesions be described
and documented by the size in centimeters of
single lesions and the cumulative sizes of multiple
lesions.206 The exact location should also be documented. OL lesions found on the floor of the mouth,
on the lateral or ventral tongue, and possibly on
the soft palate have been thought to have a greater

Vital tissue staining has been identified as an

adjunct to the early recognition of malignant or
premalignant oral lesions.211 The stain of choice is
toluidine blue (tolonium chloride). The toluidine
blue stain is a metachromatic dye that stains mitochondrial DNA, cells with an abnormally high DNA
content, and altered DNA in atypical and malignant cells. This staining technique, although very
effective, is best used in the hands of experienced
persons and at institutions where a large number of
patients with head and neck cancer are seen. When
used in these settings, this staining assists in identifying sites of malignant change and possible highgrade dysplasia. It can also be helpful in selecting
biopsy sites and in outlining lesion margins.211
Exfoliative cell collection is another, newer technique in which exfoliated oral mucosal cells are
obtained for cytologic and molecular analysis. The
newer types of exfoliative cell collection allow one
to collect more full-thickness epithelial samples,
which aids in diagnostic accuracy. Nonetheless, it is
not a highly accurate adjunctive technique at this

Etiology and Pathogenesis
The etiology and pathogenesis of OE are not well
understood. It is still unclear whether or not OE
develops de novo or from a preexisting lesion of
OL. The greatest risk factors for its development
are thought to be tobacco and alcohol use. It is

Copyright McGraw-Hill Companies, Inc. All rights reserved.

Chapter 113:

not known precisely how alcohol and tobacco use

together induce carcinoma, but the synergistic
effect of the two substances in causing intraoral
carcinoma is well accepted.
The largest OE series in the literature was in a
study conducted in Kerala, India.218 This study
demonstrated a dramatically increased risk of OE in
adults that was associated with chewing tobacco
in particular, tobacco in the form of betel quid and
pan masala. This randomized, population-based
screening trial confirmed 100 cases of OE among
49,174 persons screened (0.2%). The greatest risk
factor for development of OE was the use of chewing tobacco, with a strong dose response associated with increasing number of years of usage and
with daily usage. The independent effects of smoking and alcohol use were smaller, but the use of
chewing tobacco combined with either smoking or
alcohol consumption conferred a greater risk than
the use of chewing tobacco alone.
This study is also of importance in the United
States, because in the last three decades there has
been an increase in the use of smokeless tobacco
products in the form of moist snuff, most notably
among adolescent and young adult males.219 A
survey of 17,000 school-aged children 1217 years
old found oral premalignant lesions in 2.9% of
males and 0.1% of females.220 In this study, the risk
of development of such lesions among white male
students was strongly associated with chewing
tobacco and using snuff. Another trend of concern
among US youth is the rise in the use of other
tobacco products, such as sweet-flavored cigarettes,
bidis (hand-rolled filterless flavored cigarettes), and
kreteks (clove cigarettes).219

Histopathologic examination of a lesion of erythroplakia usually reveals findings of SCC in situ, areas
of dermal invasion, or various degrees of dysplasia.
Partial or sampling biopsies of erythroplakia may
unfortunately miss focal areas of invasive SCC. In
one study of clinically homogeneous OE, histopathologic examination revealed that 51% of cases
demonstrated invasive carcinoma, 40% showed severe dysplasia or SCC in situ, and 9% demonstrated
mild-to-moderate dysplasia.221

Epithelial Precancerous Lesions 175

Diagnosis and Differential

OE is a diagnosis of exclusion. The onus is upon
the clinician to rule out all other erythematous oral
lesions before the term oral erythroplakia (OE) can
be applied. Other entities that should be considered in the differential diagnosis include acute and
chronic mechanical trauma, thermal or chemical
injury, erythematous candidiasis, atrophic lichen
planus, lupus erythematosus, pemphigus, cicatricial pemphigoid, Kaposi sarcoma, amelanotic
melanoma, chronic contact or allergic dermatitis,
submucosal hemorrhage, and various forms of
glossitis (see Box 113-5). As OE is usually solitary,
this helps to distinguish it from other inflammatory
conditions, which are often bilateral or multiple. Of
all of these possibilities, erythematous candidiasis
and atrophic lichen planus are the most important
considerations. In general, biopsy of any OE lesion
should be performed if the cause is not obvious, if
the lesion is in a high-risk location in the oral cavity,
or if the lesion does not resolve over a period of
several weeks.

Prognosis and Clinical Course

OE has the greatest risk for malignant transformation of all premalignant oral lesions. This elevated
risk for malignant transformation is mostly based
on the fact that histopathologic evaluation of OE
usually reveals invasive SCC, SCC in situ, or severe
dysplasia, as previously discussed. The rate at which
in situ carcinoma of the oral cavity progresses to
invasive SCC is not known exactly. The transformation rate for OSCC in situ or lesions with severe
dysplasia has ranged from 14% to 50%, with an
average of 26%.216 In individuals at high risk, such
as heavy smokers or drinkers, up to 80% of OE lesions on biopsy may contain focal areas of invasive

Treatment and Prevention

Because OE has been shown to have the highest
risk for malignant transformation, early and effective treatment of these lesions is imperative. The
definitive treatment is controversial, and studies of
the treatment of OE lesions by themselves are lacking. Recurrence rates after treatment of OE lesions
are even less clear.

Copyright McGraw-Hill Companies, Inc. All rights reserved.

176 Chapter 113: Epithelial Precancerous Lesions

For severely dysplastic or in situ carcinoma lesions of erythroplakia, surgical excision or Mohs
micrographic surgery is generally recommended,
because the entire lesion can be removed with the
most thorough histopathologic confirmation of adequate margins. The proper management of mildly
or moderately dysplastic OE lesions is less clear,
with some advocating excision and others promoting a more conservative, observational approach.
Other treatment modalities, such as laser ablation,
cryosurgery, and electrodesiccation, have also been
used, but their roles in treating OE are even less
clearly defined. Given the high malignant transformation rate of these lesions, it is difficult to advocate any nonexcisional ablative therapy. Regardless
of the treatment method used to remove the lesion,
all patients should be followed at regular intervals
to be evaluated for recurrence or for the development of a not uncommon second primary lesion in
the oral cavity or aerodigestive tract. Exposure to
carcinogenic stimuli, such as tobacco and alcohol,
should be discontinued.

Copyright McGraw-Hill Companies, Inc. All rights reserved.