Basal cell carcinoma: Epidemiology, pathogenesis, clinical features, and diagnosis

INTRODUCTION — Basal cell carcinoma (BCC) is a common skin cancer arising from the basal layer of epidermis and its appendages. These tumors have been referred to as "epitheliomas" because of their low metastatic potential. However, the term carcinoma is appropriate since they are locally invasive, aggressive, and destructive of skin and the surrounding structures, including bone (picture 1A-B).

The epidemiology, pathogenesis, clinical presentation, and differential diagnosis of BCC will be reviewed here. The treatment and prognosis of BCC are discussed separately. Gorlin syndrome is also discussed separately. (See "Treatment and prognosis of basal cell carcinoma at low risk of recurrence" and "Treatment of basal cell carcinomas at high risk for recurrence" and "Nevoid basal cell carcinoma syndrome (Gorlin syndrome)".)

EPIDEMIOLOGY — Estimates of the incidence of BCC are imprecise since in most countries there is no cancer registry that collects data on BCC [1]. The American Cancer society estimates that in 2012, 5.4 million cases of nonmelanoma skin cancers (NMSCs) were diagnosed in 3.3 million people, of which approximately 8 in 10 cases would have been BCC [2]. A population-based study estimated that 3.5 million NMSCs were treated in the United States in 2006 [3]. One study using data from a commercially insured population in the United States estimated an age-adjusted incidence and prevalence of BCC of 226 and 343 per 100,000 persons per year, respectively [4]. Similar rates were reported in a population-based study in Olmsted County, Minnesota [5], while higher rates were estimated in the Nurses' Health Study (1986 to 2006) and Health Professionals Follow-up Study, where the age-adjusted BCC incidence rates were 519 cases per 100,000 person-years in females and 606 cases per 100,000 person-years in males. Over a 20-year follow-up period, the incidence of BCC increased to 1019 cases per 100,000 person-years and 1488 cases per 100,000 person-years in female and male patients, respectively [6].

Multiple epidemiologic observations provide insights into the etiology of BCC:

●BCC is particularly common in White populations; it is very uncommon in people with darkly pigmented skin. In White populations in the United States, the incidence of BCC has increased by more than 10 percent per year, and the lifetime risk of developing a BCC is 30 percent [7].

●An increasing incidence of BCC over time has been noted in the United States and other countries, including Canada, Finland, and Australia [6,8-10]. Of greater concern, there may be an increasing incidence of aggressive-growth histologic subtypes, which are more difficult to treat [11].

●The incidence in men is 30 percent higher than in women, particularly with the superficial type [8,12,13].

●Within the United States, there is striking geographic variation in incidence. States closer to the equator, such as Hawaii and California, have an incidence of BCC at least twice that of the Midwestern United States [13,14].

●There are also prominent global variations in incidence. Northern European countries, such as Finland, have an incidence one-fourth that of the Midwestern United States. This contrasts with Australia where rates are 40 times that of Finland [8,9,12].

●The incidence of BCC increases with age; persons aged 55 to 75 have about a 100-fold higher incidence of BCC than those younger than 20 [15]. Although increasing longevity may underlie some of the increasing incidence of BCC, the incidence of BCC among Americans younger than 40 also appears to be increasing, particularly among women [16].

RISK FACTORS — Environmental, phenotypic, and genetic factors contribute to the development of BCC. Although exposure to ultraviolet (UV) radiation in sunlight is the most important risk factor for BCC, other established risk factors include chronic arsenic exposure, radiation therapy, long-term immunosuppressive therapy, and nevoid basal cell carcinoma syndrome (NBCCS; Gorlin syndrome).

Environmental

Ultraviolet radiation

Sun exposure — Exposure to UV radiation from sunlight is the most important environmental cause of BCC, and most risk factors relate directly to a person's sun exposure habits or susceptibility to solar radiation. These risk factors include having fair skin, light-colored eyes, red hair, northern European ancestry, older age, childhood freckling, and an increased number of past sunburns [17-19].

The type, quantity, and timing of sun exposure associated with an increased risk of BCC are not clearly defined. Childhood sun exposure appears to be more important than exposure during adult life [17,20]. Evidence supporting this hypothesis comes from case control studies and clinical trials [18-21].

In a Canadian, case-control study that included 226 men with BCC and 406 age-matched controls, the development of BCC was strongly correlated with childhood and adolescent sun exposure but not cumulative or recent sun exposure [20]. In other studies, however, adult sun exposure was a risk factor for BCC [18].

The frequency and intensity of sun exposure may also be important. Solar exposure in intermittent, intense increments increases the risk of BCC more than a similar dose delivered more continuously over the same period of time [22]. A French case-control study including more than 1000 women with BCC and more than 3600 controls found that a history of severe sunburn before the age of 25 years and after the age of 25 were both independently associated with a twofold increased risk of BCC, after adjusting for skin sensitivity to sunlight and hair, eye, and skin color [23]. The role of chronic occupational sun exposure remains uncertain [24].

Tanning beds — The use of tanning beds may increase the risk for early development of BCC [25-27]:

●A cohort study of approximately 73,000 female nurses found that women who used tanning beds more than six times per year during high school or college were more likely to develop BCC than women who did not use tanning booths during these time periods (adjusted hazard ratio 1.73, 95% CI 1.52-1.98) [28].

●A 2012 meta-analysis of 12 observational studies (9328 nonmelanoma skin cancer [NMSC] cases) found that subjects with a history of any tanning bed use were more likely to develop BCC than those who had never used tanning beds (relative risk [RR] 1.29, 95% CI 1.08-1.53). The risk was slightly higher for those reporting their first use of tanning devices before the age of 25 (RR 1.40, 95% CI 1.29-1.52) [29].

●A population-based, case-control study that included approximately 650 cases of BCCs and 450 controls found that tanning bed users had a 60 percent increased risk of developing a BCC at or before the age of 50 years (odds ratio [OR] 1.6, 95% CI 1.3-2.1) [26]. In this study, the risk of BCC was doubled for those reporting their first use of tanning devices before the age of 20.

●A Canadian study found that the RR of BCC associated with ever using tanning beds was 1.39 (95% CI 1.10-1.76) and estimated that 5 percent of BCCs were attributable to ever use of indoor tanning devices [30].

Tanning bed use may be a marker of populations more exposed to the sun. Studies have shown that tanning bed users are more likely to be regular sunbathers and to have poorer sun-protection behavior than nonusers [31,32]. In these epidemiologic studies, the amount of ultraviolet A (UVA) versus ultraviolet B (UVB) and other wavelengths of nonionizing radiation to which participants were exposed in tanning beds is not known.

Phototherapy — Therapeutic exposure to psoralen plus ultraviolet A light (PUVA) for cutaneous disorders such as psoriasis increases the risk of NMSC, particularly squamous cell carcinoma (SCC) [33]. The risk of BCC in patients treated with PUVA is lower than the risk for cutaneous SCC. In a 30-year prospective cohort study documenting the incidence of NMSC in patients given PUVA for psoriasis, the increase in risk for BCC was modest compared with cutaneous SCC [34]. (See "Psoralen plus ultraviolet A (PUVA) photochemotherapy", section on 'Skin cancer'.)

Broadband UVB (280 to 320 nm) and narrowband UVB (311 to 313 nm) phototherapy appear to be less likely to promote the development of NMSC than PUVA [35], but further research is necessary to determine the true carcinogenic potential of these therapies. Most studies examining the role of UVB have examined narrowband UVB alone or broadband UVB in combination with narrowband UVB [35-41]. Although some observational studies have reported a small increased risk for BCC after UVB therapy [36,37], others have reported no increased risk [38-41]. As an example, in a retrospective study of 3867 phototherapy patients (of which 352 received ≥100 narrowband UVB treatments), narrowband UVB was not associated with an increased risk of BCC, except in cases where patients were treated with both narrowband UVB and PUVA [38]. (See "UVB phototherapy (broadband and narrowband)".)

Photosensitizing agents — The association of BCC with UV light exposure has led to questions about the impact of photosensitizing drugs on the development of BCC. An association between prior use of photosensitizing tetracyclines, calcium channel blockers, beta blockers, or thiazide diuretics and a modest increase in risk for BCC has been documented in observational and case-control studies [42-47]:

●A meta-analysis of eight observational studies with nearly 400,000 participants found a weak association between the use of thiazide diuretics and the risk of BCC (OR 1.19, 95% CI 1.02-1.38) [48].

●In addition, an American, population-based, case-control study of 1567 adults with BCC and 1906 controls found a modest increase in risk for multiple BCC (OR 1.4, 95% CI 1.0-2.1) and BCC before the age of 51 (OR 1.5, 95% CI 1.1-2.1) among participants who recalled use of a photosensitizing medication [42].

Beyond drugs, dietary consumption of photosensitizing agents may contribute to the risk of developing BCC. A study using participants in the Health Professionals Follow-up Study and the Nurses' Health Study examined the effect of dietary intake of photocarcinogenic agents, such as furocoumarins in the form of citrus products, on NMSC [49]. Compared with those who consumed citrus products less than twice per week, the pooled, multivariable-adjusted hazard ratios for BCC increased in a dose-dependent fashion up to 1.16 (95% CI 1.09-1.23) for citrus consumption 1.5 times per day or more, supporting an association between high citrus consumption and a slightly increased risk of BCC in those cohorts.

Additional studies are necessary to clarify the relationship between photosensitizing agents and BCC. (See "Photosensitivity disorders (photodermatoses): Clinical manifestations, diagnosis, and treatment", section on 'Photosensitivity due to exogenous agents' and "Cutaneous squamous cell carcinoma: Epidemiology and risk factors", section on 'Thiazide diuretics and other photosensitizing drugs'.)

Chronic arsenic exposure — Superficial multicentric BCC may occur due to chronic exposure to arsenic from ingestion of contaminated drinking water, seafood, or medications [50-54]. (See "Arsenic exposure and chronic poisoning", section on 'Cancer'.)

In a nationwide population-based study using registry data from the National Taiwan Cancer Registry Center between 1979 and 2007, the risk of BCC was three- to fourfold higher in the blackfoot disease, a form of peripheral vascular disease associated with arsenic exposure, endemic areas compared with other areas of Taiwan [53].

The risk of BCC associated with arsenic exposure may be influenced by genetic factors, such as variants of the AS3MT gene, encoding the arsenite methyltransferase enzyme, and telomer length [54,55]. In a study including 528 arsenic-exposed cases with BCC and 533 healthy controls, within each tertile of arsenic exposure, individuals with shorter telomeres were at increased risk of BCC, with the highest risk in the highest exposure group [55].

Ionizing radiation — Superficial therapeutic ionizing radiation, as for facial acne, psoriasis, or tinea capitis, increases the risk of NMSC, including BCC [56-58]. The latency period for development of BCCs is about 20 years, and lesions are limited to sites within the radiation field.

The treatment of noncutaneous disorders with radiation therapy (eg, thymic enlargement in infancy and ankylosing spondylitis) as well as the use of radiation therapy for conditioning prior to hematopoietic stem cell transplantation have also been associated with the appearance of BCC, particularly with exposure at a younger age [57,59,60].

Ionizing radiation used to treat childhood cancers also increases the risk for the subsequent development of BCC. This was illustrated in a study of 776 subjects, five of whom developed BCCs, approximately 10-fold more than was expected in this population [61]. All of the BCCs were located within the radiation field. In addition, a case-control study of 199 patients with a history of both childhood cancer and BCC and 597 controls with a history of childhood cancer without BCC found a linear dose-response relationship between the radiation dose and risk for BCC [62]. An increase in risk was detected among patients who received at least 1 Gy of radiation to the skin, and patients who received 35 Gy or more were 40 times more likely to develop BCC than those who were not treated with radiation (OR 39.8, 95% CI 8.6-185). Occupational exposure to ionizing radiation (ie, as a radiation technologist) may also increase the risk of BCC development [63].

BCC development is strikingly absent in Black survivors of irradiated childhood cancer [64] and substantially lower in Black versus White patients who received radiation for tinea capitis [65] for reasons that are not understood or completely explained by darker skin pigmentation alone [64].

Studies of survivors of the atomic bomb explosions in Japan support the role of exposure to ionizing radiation in the development of BCC in the nonmedical setting [66-69].

Phenotypic traits — Light skin pigmentation, light hair and eye color, and poor tanning ability, which reflect the skin sensitivity to sunlight, are well-known risk factors for BCC. A meta-analysis of 29 observational studies found that red hair, fair complexion, and skin that burns but never tans were associated with a twofold risk of developing a BCC [70].

Personal history of basal cell carcinoma — Individuals with a history of BCC are at increased risk for subsequent lesions. Approximately 40 to 50 percent of patients who have had one BCC will develop another lesion within five years [71,72]. However, the probability of developing a subsequent BCC is significantly less after a first BCC than after a nonfirst BCC (13 versus 34 percent at one year, 20 versus 52 percent at two years, and 35 versus 75 percent at five years) [73]. (See "Treatment of basal cell carcinomas at high risk for recurrence", section on 'Prognosis and follow-up'.)

Predisposing genetic variants — In addition to specific mutational drivers of BCC (see 'Pathogenesis' below), germline polymorphisms in genes that determine pigmentary traits, such as melanocortin-1 receptor (MC1R), human homolog of agouti signaling protein (ASIP), and tyrosinase (TYR), are associated with increased risk of BCC [74-76]. However, independent of melanocortin 1 receptor (MCR1) phenotype, a family history of skin cancer is associated with increased risk of BCC under age 40 (OR 2.49, 95% CI 1.80-3.45) [77].

Specific gene polymorphisms have been associated with the truncal phenotype and clustering of BCCs. Patients are often younger, male, and have more clusters of BCCs compared with those with BCCs arising in sun-exposed sites. The associated genes include those encoding the detoxifying enzyme genes cytochrome P-450 CYP2D6 and glutathione S-transferase, the vitamin D receptor, and tumor necrosis factor [78-82]. The link between these genetic polymorphisms, tumorigenesis, and the clinical phenotype is not clear.

Genome-wide association studies have identified genetic variants that may influence BCC risk through other pathways, such as an effect on the growth or differentiation of basal layers of the epidermis or an effect on the TP53 tumor suppressor gene [83-86]. A genome-wide association meta-analysis found that single nucleotide polymorphisms in genes involved in DNA excision repair may be involved in the pathogenesis of BCC [87].

Genes that affect the immune response may also impact susceptibility to BCC. Cytotoxic lymphocyte-associated antigen-4 (CTLA-4) is expressed on regulatory T cells and is involved in UV-induced immune tolerance. In a case-control study, genetic variation at the CTLA4 locus influenced the risk of BCC, particularly among patients with a higher number of severe sunburns [88].

Inherited disorders — Inherited disorders that are associated with a greatly increased risk of developing BCCs at an early age and with increased morbidity include:

●Nevoid basal cell carcinoma syndrome – Nevoid basal cell carcinoma syndrome (NBCCS), also known as basal cell nevus syndrome or Gorlin syndrome, is a rare multisystem disorder of autosomal dominant inheritance caused in most cases by germline mutations of the human patched gene (PTCH1) [89]. Affected patients have both developmental anomalies and postnatal tumors, including multiple BCCs, at an average age of 20 to 21 years, odontogenic keratocysts, and medulloblastoma [90]. (See "Nevoid basal cell carcinoma syndrome (Gorlin syndrome)" and 'PTCH1 mutations' below.)

●Rombo syndrome – Rombo syndrome was first described in a family with vermiculate atrophoderma and peripheral vasodilation with cyanosis in childhood, milia, trichoepitheliomas, hypotrichosis in adulthood, and BCCs developing in the third and fourth decade [91]. Rombo syndrome appears to be transmitted in a dominant pattern; however, a causative mutation had not been identified.

●Bazex-Dupré-Christol syndrome – Bazex-Dupré-Christol syndrome (also called Bazex syndrome or follicular atrophoderma and basal cell carcinomas) is an X-linked dominant disorder characterized by congenital hypotrichosis, follicular atrophoderma, milia, and multiple BCCs [92].

●Xeroderma pigmentosum – Xeroderma pigmentosum is a rare, autosomal recessive disorder due to mutations in any of eight genes involved in repair of UV-induced DNA damage [93]. Clinical findings include early-onset pigmentary skin changes and early development of skin cancers. SCCs and BCCs develop at an average age of nine years. (See "Xeroderma pigmentosum".)

●Muir-Torre syndrome – Muir-Torre syndrome is a rare, autosomal dominant condition caused by mutations in DNA mismatch repair genes MLH1, MSH2, and MSH6. Patients present with sebaceous neoplasms, including sebaceous adenomas and carcinomas, keratoacanthomas, BCCs, and malignancies of the colon and genitourinary tract [91]. (See "Muir-Torre syndrome".)

●Oculocutaneous albinism – Oculocutaneous albinism (OCA) is a group of autosomal recessive disorders of melanin biosynthesis presenting with a spectrum of visual disturbances and hypopigmentation of the skin and hair. Individuals with OCA have an increased risk of early-onset skin cancer, possibly by their teenage years. SCC is the most common type of cancer occurring in patients with OCA, but BCC and melanoma also occur [94]. (See "Oculocutaneous albinism".)

Immunosuppression — Chronic immunosuppression (as occurs with solid organ transplantation and with HIV infection) may increase risk for the development of BCC, although the increase in risk is less than that observed for SCC [95,96]. The risk for BCC after solid organ transplantation appears to increase linearly over time, whereas the risk for SCC rises exponentially [95]. As in other populations, sun exposure, phenotype, and other factors influence the likelihood that an organ transplant recipient will develop BCC [97]. (See "Epidemiology and risk factors for skin cancer in solid organ transplant recipients", section on 'Squamous cell and basal cell carcinoma'.)

The increased risk for skin cancer in organ transplant recipients is attributed to chronic exposure to immunosuppressive agents [98,99]. The specific impact of systemic glucocorticoid therapy on BCC risk is uncertain; studies performed in patients without a history of organ transplantation conflict on whether systemic glucocorticoid therapy significantly increases risk for BCC [100-103].

There is increased risk of skin cancer in the allogeneic stem cell transplant population as well. A Danish study of 1007 patients who received allogeneic stem cell transplants found a hazard ratio for BCC of 3.1 (95% CI 1.9-5.2), compared with the background population. This rate was on par with the renal transplant recipients [104]. Another study of adult allogeneic stem cell transplant recipients at two Boston hospitals reported an incidence rate ratio of 2.5 for BCC development (95% CI 1.9-3.2) [105]. The reasons for the increase are not clear and have been attributed in part to the conditioning regimen (including total body irradiation), disease prior to transplant (ie, chronic lymphocytic leukemia), and presence of graft-versus-host disease [104-106].

Less is known about the risk for skin cancer in nontransplanted patients treated with immunosuppressants other than glucocorticoids. In a retrospective cohort study of 405 patients with rheumatoid arthritis (n = 349, 86 percent) or psoriatic arthritis (n = 56, 14 percent), the use of methotrexate and methotrexate with cyclosporine A or D-penicillamine was associated with an increased risk of NMSC [107]. Among patients treated with methotrexate, a dose-response relationship was noted only for BCC, with a standardized incidence ratio of 5.77 (95% CI 4.19-7.74) for those treated with a cumulative dose >8 grams versus 2.21 (95% CI 1.35-3.41) for those treated with a cumulative dose <5 grams. However, these results should be interpreted with caution as the possibility of ascertainment bias, due to increased medical surveillance for patients with rheumatoid arthritis or psoriatic arthritis, cannot be excluded.

Support for an increased risk for BCC among HIV-positive individuals was demonstrated in a retrospective cohort study of HIV-positive (n = 6560) and HIV-negative (n = 36,821) non-Hispanic White patients in a northern California healthcare system. The study found that patients with HIV infection were approximately twice as likely to develop BCC as patients without HIV infection (adjusted rate ratio 2.1, 95% CI 1.8-2.3) [96].

Other factors — Other factors that have been associated with an increased risk of BCC include:

●Nevus sebaceous – Nevus sebaceous is an uncommon congenital hamartoma of the skin composed of epidermal, follicular, sebaceous, and apocrine elements. BCC may develop within nevus sebaceous, though this occurrence is rare. In a retrospective study of 596 nevus sebaceous excisions from adults and children, BCC was detected in specimens from five adults (0.8 percent) [108]. A separate review of 631 children with 651 lesions of nevus sebaceous found that BCC may also develop within nevus sebaceous in children; BCC was found in excisional specimens from five patients (0.8 percent) [109]. (See "Nevus sebaceus and nevus sebaceus syndromes".)

●Lifestyle factors – Smoking increases the risk of SCC [72] and has been evaluated for an association with BCC [21,110]. Although a case-control study performed in 333 disease-discordant twin pairs found an increased risk of BCC in smokers and in females compared with males [110], a subsequent meta-analysis failed to find a significant association between BCC and smoking [111]. A small, dose-dependent increase in BCC risk of 1.07 (1.04-1.09) has been shown with every 10 gram/day increase in ethanol consumption [112]. Other lifestyle factors possibly affecting the risk for BCC include higher education and coffee consumption [110,113,114]. Additional studies are needed to confirm these findings before firm conclusions can be drawn.

●Human papillomavirus – Although an epidemiologic and biologic relationship between human papillomavirus (HPV) and BCC has not been established, some studies in organ transplant patients and psoriasis patient populations have noted an increased number of NMSCs associated with evidence of HPV in the skin [115,116].

PATHOGENESIS

Ultraviolet radiation-induced carcinogenesis — Mutations in several tumor suppressor genes and proto-oncogenes have been implicated as drivers in BCC formation [117]. In almost 90 percent of cases, alterations of genes causing hyperactivation of the hedgehog (HH) protein family, a highly conserved developmental pathway involved in organogenesis, tissue repair, and including PTCH1 receptor, SMO signal transducer, and GLI transcription factors, are linked to BCC formation. Beyond the HH pathway, the TP53 tumor suppressor gene is also commonly implicated in the pathogenesis of BCC.

Within these genes, the mutational profiles of BCCs reveal evidence of ultraviolet (UV) radiation-induced mutagenesis. In particular, mutations identified in PTCH1 and TP53 are consistent with UV radiation-induced mutagenesis in most cases. These so-called UV signature mutations include base substitutions of C>T at dipyrimidine sites and CC>TT tandem base substitutions, although the latter occur less frequently [118]. This is true for BCCs that arise sporadically, and even more so for BCC arising in patients with xeroderma pigmentosum, suggesting that repair of UV-induced DNA damage can reduce BCC carcinogenesis. Beyond UV-induced changes, other factors, such as oxidative stress, are also associated with the mutagenesis of BCCs [117,119]. (See "Xeroderma pigmentosum".)

PTCH1 mutations — Our understanding of the pathogenesis of BCC was greatly enhanced with the discovery of mutations in the PTCH1 gene on chromosome 9q22.3 in patients with inherited nevoid basal cell carcinoma syndrome (NBCCS) [120] (see "Nevoid basal cell carcinoma syndrome (Gorlin syndrome)"). Subsequently, somatic mutations of PTCH1 were identified in over 70 percent of sporadic BCCs and BCCs associated with xeroderma pigmentosum, indicating that an aberrant HH signaling activation is a prerequisite for the development of both BCC associated with the Gorlin syndrome and sporadic BCCs [117,119,121-125]. In a manner similar to the retinoblastoma gene, two somatic "hits" in the same cell are required for sporadic cases, while one somatic "hit" plus the inheritance of one defective allele is responsible for the familial cases.

PTCH1 encodes a protein acting as a transmembrane receptor for the HH protein family, an important component of the HH signaling pathway, which directs embryonic development of a variety of organs in vertebrates (figure 1) [126]. Three HH ligands are present in mammals: sonic hedgehog (SHH), Indian hedgehog (IHH) and desert hedgehog (DHH). SHH is the most studied HH ligand; it binds a cell membrane receptor complex that is formed by PTCH and a second protein, smoothened (SMO), relieving the inhibition of the pathway induced by unbound PTC1 and thus activating the HH pathway [121,127]. However, the mechanism by which HH signal overexpression leads to tumorigenesis is unclear. It may involve the activation of the transcription factors Gli1 (glioma-associated oncogene homolog) and/or Gli2.

TP53 mutations — The second most important gene in BCC carcinogenesis is TP53, encoding the P53 protein involved in maintaining genomic stability by regulating the cell cycle, inducing apoptosis, and activating DNA repair. TP53 mutations have been detected in 20 to over 60 percent of sporadic BCCs [117].

Other genes — In addition to PTCH1 and TP53, other tumor-related genes have been implicated in BCC pathogenesis [117-119,128]. In a series of 293 BCCs, 85 percent harbored mutations in genes involved in the HH pathway (PTCH1 73 percent, SMO, 20 percent, and SUFU 8 percent), and TP53 (61 percent) and also in multiple other cancer-related genes, such as MYCN, PPP6C, PTPN14, STK19, and LATS1 [119]. In a whole-exome sequencing study of 12 sporadic BCCs and normal skin, mutations were found in a number of known or putative cancer genes, including CSMD1, DPP10, NOTCH1, and PREX2; mutational hotspots were detected in STAT5B, CRNKL1, and NEBL [118]. However, the relevance of these mutations in the BCC pathogenesis is unclear.

PATHOLOGY — On histopathologic examination, common findings of BCC are nodules and/or strands of atypical basaloid cells that show nuclear palisading, cellular apoptosis, and scattered mitotic activity in the dermis (picture 2A-B). Artifactual cleft formation may be seen between the tumor lobules and the surrounding stroma, which may be mucinous. Solar elastosis is usually present in the dermis. The histologic patterns of BCC (nodular, superficial, morpheaform/infiltrative, basosquamous, micronodular, and pigmented) are often reflected in the clinical appearance:

●Hematoxylin-eosin staining of nodular BCC, the most common subtype, reveals discrete nests of basaloid cells in the dermis. There is peripheral palisading of the malignant keratinocytes and a mucinous-surrounding tumor stroma. Between the tumor and the dermis, a separation or "cleft," is often apparent due to retraction artifact in tissue processing.

●Superficial BCC is characterized histologically by foci of malignant, basaloid, palisading tumors "budding" off the epidermis.

●In morpheaform/infiltrative BCCs, there are thin cords of basaloid tumor cells penetrating the surrounding collagen, which may appear sclerotic.

●Micronodular BCCs appear as numerous small collections of malignant basaloid cells within the dermis and exhibit more subtle findings of peripheral palisading and retraction compared with nodular BCCs.

●Pigmented BCCs represent approximately 6 percent of BCCs and are so named due to melanin and melanophages found within the tumor nodules.

●Basosquamous BCCs have a component of nodular or superficial BCC overlying an invasive portion that has features of BCC and squamous cell carcinoma (SCC) [129,130].

●A less common and indolent subtype, fibroepithelial BCCs (fibroepitheliomas of Pinkus), exhibit thin strands of basaloid keratinocytes in a reticular pattern and a spindle celled stroma.

Morpheaform/infiltrative, micronodular, and basosquamous are considered more "aggressive growth" subtypes of BCC [11]. Some lesions have a mixed histology, and up to 40 percent have features of more than one histologic subtype [130,131].

CLINICAL PRESENTATION — Approximately 70 percent of BCCs occur on the face, consistent with the etiologic role of solar radiation, and 15 percent present on the trunk. Only rarely is BCC diagnosed on areas like the penis, vulva, or perianal skin [132]. The clinical presentation of BCC can be divided into three groups, based upon lesion histopathology: nodular, superficial, and morpheaform.

Nodular — Nodular BCCs, which represent approximately 80 percent of cases, typically present on the face as a pink or flesh-colored papule (picture 3G) [133]. The lesion usually has a pearly or translucent quality, and a telangiectatic vessel is frequently seen within the papule. The papule may often be described as having a "rolled" border, where the periphery is more raised than the middle. Ulceration is frequent (picture 3J), and the term "rodent ulcer" refers to these ulcerated nodular BCCs (picture 1A-C).

Superficial — Approximately 15 percent of BCCs are superficial BCCs [133]. For unclear reasons, men have a higher incidence of superficial BCC than do women.

Superficial BCCs most commonly occur on the trunk and typically present as slightly scaly, nonfirm macules, patches, or thin plaques light red to pink in color (picture 3A-F). The center of the lesion sometimes exhibits an atrophic appearance, and the periphery may be rimmed with fine translucent papules. A shiny quality may be evident when a superficial BCC is illuminated. Occasionally, spotty brown or black pigment is present, which may contribute to confusion with melanoma (picture 3D).

Superficial BCCs tend to grow slowly and can vary in size from macules measuring just a few millimeters in diameter to lesions several centimeters in diameter or more if left untreated. Superficial BCCs are usually asymptomatic.

Morpheaform/infiltrative — Morpheaform or sclerosing BCCs constitute 5 to 10 percent of BCCs. These lesions are typically smooth, flesh-colored, or very light pink papules or plaques that are frequently atrophic; they usually have a firm or indurated quality with ill-defined borders (picture 4A-B). Infiltrative and micronodular subtypes are less common than the morpheaform BCC.

Other subtypes — Several other BCC subtypes have been described. Basosquamous cell carcinoma is a rare tumor that behaves aggressively. Both nodular and superficial BCCs can produce pigment. These lesions are referred to as pigmented BCCs (picture 3H). (See "Treatment and prognosis of low-risk cutaneous squamous cell carcinoma (cSCC)".)

Multiple basal cell carcinoma syndromes — Several rare syndromes have been described that present with multiple BCCs. The most common is nevoid basal cell carcinoma syndrome (NBCCS). (See "Nevoid basal cell carcinoma syndrome (Gorlin syndrome)".)

Patients with xeroderma pigmentosum and Muir-Torre syndrome are at increased risk for BCC as well as other skin cancers. The incidence of BCCs, squamous cell carcinomas (SCCs), and melanomas for individuals with xeroderma pigmentosum under the age of 20 is approximately 2000 times greater that seen in the general population. (See "Xeroderma pigmentosum" and "Muir-Torre syndrome".)

Natural history — Most BCCs remain localized, and the growth rate is variable. However, a few become locally aggressive or metastatic, and the acquisition of cytogenetic aberrations may be associated with aggressive biologic behavior. In one series, for example, trisomy 6 was identified in none of 22 nonaggressive, two of four locally aggressive, and all four metastatic BCCs [134].

DIAGNOSIS

Clinical and dermoscopic examination — Clinicians who are familiar with the clinical manifestations of BCC are often able to make the diagnosis based upon clinical examination (picture 3A-B, 3D, 3G-J). Examination of the lesion with a dermatoscope may assist in the clinical diagnosis of BCC [135-137].

Dermoscopic features of BCC include the lack of a pigmented network (which is typically associated with melanocytic lesions) and the presence of one or more findings that are characteristic of BCC, such as arborizing vessels, blue-gray ovoid nests, and ulceration (figure 2 and picture 5). A meta-analysis of 13 studies found that the pooled sensitivity and specificity of dermoscopy for the diagnosis of BCC were 91 and 95 percent, respectively; in a subgroup of five studies comparing the accuracy of naked eye examination followed by dermoscopy with naked eye examination alone, the sensitivity and specificity were 85 percent and 98 percent, respectively [138]. (See "Dermoscopic evaluation of skin lesions", section on 'Criteria for basal cell carcinoma'.)

However, a skin biopsy is usually performed to provide pathologic confirmation of the diagnosis and determine the histologic subtype.

Biopsy — In cases for which the clinical diagnosis of BCC appears certain and the tumor lacks clinical features associated with a high risk for tumor recurrence, clinicians experienced in the diagnosis of BCC sometimes elect to perform the biopsy at the same time as definitive treatment (eg, immediately prior to electrodesiccation and curettage). In addition, some clinicians choose to treat lesions without a biopsy when high-risk clinical features are absent and the patient has a history of multiple similar low-risk BCCs. (See "Treatment of basal cell carcinomas at high risk for recurrence", section on 'Features associated with high risk for recurrence' and "Treatment and prognosis of basal cell carcinoma at low risk of recurrence".)

However, the decision not to perform a biopsy prior to definitive treatment is not without risk. Because the histologic features of a tumor provide additional information on the risk for tumor recurrence following treatment, not performing a biopsy prior to definitive treatment may result in a failure to detect a tumor with aggressive histologic features that might be best managed with a different approach to therapy. The misdiagnosis of a different tumor as BCC (eg, amelanotic melanoma) is an additional risk of deferring a biopsy (picture 6A-B).

To reduce the risk for patient mismanagement, we suggest always performing a biopsy in the following situations:

●The lesion exhibits features atypical for BCC

●The patient lacks a prior history of BCC

●The lesion exhibits clinical features suggestive of a BCC with a high-risk for recurrence

Shave biopsies, punch biopsies, and excisional biopsies can be used for the diagnosis of BCC. Although shave and punch biopsies are frequently performed for diagnosis due to the simplicity of these procedures, clinicians should be aware that biopsies that remove only a portion of the lesion do not always provide an accurate assessment of the histologic subtype of a tumor [139-143]. With punch biopsies, an aggressive histologic subtype may be missed in up to 20 percent of cases [139-141]. A retrospective study in which the majority of biopsy procedures were shave biopsies (230 shave biopsies and 27 punch biopsies) found that an aggressive histologic subtype was missed in 7 percent of cases [142]. (See "Skin biopsy techniques", section on 'Biopsy techniques'.)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis varies with the subtype of BCC (ie, nodular, superficial, or morpheaform):

●Early nodular variants with little ulceration clinically may be identical to benign growths such as dermal nevi (picture 7), small epidermal inclusion cysts, or even sebaceous hyperplasia (picture 8). A single lesion of molluscum contagiosum has a similar appearance, as does amelanotic melanoma.

●Larger lesions with central ulceration can appear cup shaped. These can resemble squamous cell carcinoma (SCC), keratoacanthomas, or dermal metastases from internal organs, such as the colon [144]. (See "Cutaneous squamous cell carcinoma (cSCC): Clinical features and diagnosis" and "Keratoacanthoma: Epidemiology, risk factors, and diagnosis".)

●Superficial BCCs may be confused with inflammatory disorders of the skin such as nummular eczema (also known as nummular dermatitis (picture 9)) or psoriasis (picture 10), especially when a peripheral rim of small, pearly papules is absent. In particular, the possibility of superficial BCC should be considered when a lesion presumed to be inflammatory fails to respond to topical corticosteroids. (See "Nummular eczema" and "Psoriasis: Epidemiology, clinical manifestations, and diagnosis".)

●Benign lichenoid keratoses (picture 11), actinic keratoses (picture 12), and rarely amelanotic melanoma (picture 6A-B) presenting as scaly erythematous macules may also be mistaken for superficial BCC. (See "Actinic keratosis: Epidemiology, clinical features, and diagnosis" and "Melanoma: Clinical features and diagnosis".)

●Morpheaform BCCs frequently appear similar to a scar or other site of trauma. The induration of the lesion simulates localized scleroderma. (See "Pathogenesis, clinical manifestations, and diagnosis of morphea (localized scleroderma) in adults".)

●Pigmented nodular or superficial BCCs may resemble melanoma or, less likely, a benign nevus. (See "Melanoma: Clinical features and diagnosis".)

PREVENTION

Sun protection — The primary approach to the prevention of BCCs is protection from sun exposure. The various techniques to minimize solar exposure are discussed elsewhere. (See "Selection of sunscreen and sun-protective measures".)

It has been estimated that aggressive sun protection before the age of 18 years could reduce the number of nonmelanoma skin cancers (NMSCs) by almost 80 percent [145]. Several trials provide evidence that sunscreen use decreases the incidence of squamous cell carcinomas (SCCs) and that there are no adverse effects from sunscreen use [146-148]. However, a randomized trial evaluating the effects of sunscreen and the antioxidant beta-carotene over a four-year period found that participants using topical sunscreen had a 40 percent reduction in SCCs but no decrease in BCCs [146].

Chemoprevention

Nonsteroidal anti-inflammatory drugs — Data conflict on the impact of nonsteroidal anti-inflammatory drugs (NSAIDs) on the risk for BCC [149-152]:

●The results of an 11-month, randomized trial suggested that celecoxib (a cyclooxygenase-2 inhibitor) may be beneficial for chemoprevention. In this trial of 240 patients with actinically damaged skin, patients treated with celecoxib (200 mg twice daily for nine months) developed fewer BCCs than patients who were given placebo (adjusted rate ratio 0.4, 95% CI 0.18-0.93) [153].

●In contrast, a Danish population-based case-control study failed to find an association between the use of celecoxib or other prescription NSAIDs and overall risk for BCC [154].

●Another population-based case-control study including over 65,000 cases with incident, first-time diagnosis of BCC and the same number of matched controls selected from the United Kingdom Clinical Practice Research Datalink did not find an association between the use of any NSAIDs and the overall risk of BCC [155]. In subgroup analyses, a modest risk reduction was observed among long-term users of ibuprofen (odds ratio [OR] 0.85, 95% CI 0.77-0.94), and risk was further reduced among mono-users (OR 0.61, 95% CI 0.48-0.78). However, these results must be interpreted with caution because of the lack of information on potential confounders or effect-modifying factors such as sun exposure and skin phototype.

Although not explicitly mentioned in those studies, adverse effects of NSAID therapy on multiple organ systems (eg, cardiovascular, gastrointestinal, renal) warrant additional consideration when recommending long-term use as chemoprevention. (See "Nonselective NSAIDs: Overview of adverse effects".)

Oral nicotinamide — The efficacy of oral nicotinamide (vitamin B3), a dietary supplement available over the counter, for the prevention of NMSC has been evaluated in a phase 3, randomized trial [156]. In this study, 386 immunocompetent participants (mean age 66 years) with ≥2 histologically confirmed NMSCs in the past five years were treated with 500 mg of nicotinamide twice daily or placebo for 12 months. The primary endpoint was the number of new NMSCs at 12 months. At the end of the study, patients in the nicotinamide group had a lower number of BCC and SCC (NMSC) than patients in the placebo group (1.8 versus 2.4), corresponding to an overall rate reduction of 23 percent (95% CI 4-38 percent) after adjustment for center and five-year NMSC history. For BCC specifically, the rate reduction was 20 percent (95% CI -6 to 0.39). There was no significant difference in adverse events between the placebo and nicotinamide groups. No effect on NMSC rates was observed in the six months after discontinuation of treatment.

An underpowered, phase 2, randomized trial of nicotinamide in 22 renal transplant patients found a statistically nonsignificant 35 and 15 percent reduction in the rate of NMSCs and actinic keratoses, respectively [157]. Further, a published response to the original phase 3 study suggested that with Bayesian reinterpretation, the 23 percent estimated reduction in NMSC would not be reproducible [158]. Additional studies are needed to evaluate the effects of long-term treatment and establish the optimal patient population and dose of medication.

Topical fluorouracil — A single course of topical fluorouracil has been shown to reduce the development of new actinic keratoses, a marker for increased risk of keratinocyte cancers (ie, BCC and SCC) in older male patients with multiple previous keratinocyte cancers [159,160]. However, topical fluorouracil seems not to be effective in preventing the development of BCC in high-risk patients.

In a randomized trial, 932 participants with a history of at least two keratinocyte cancers in the previous five years were instructed to apply topical fluorouracil 5% cream or vehicle cream twice daily to the face and ears for four weeks (a total of 56 doses) [161]. The primary study endpoints were surgically treated BCC or SCC on the face or ears and found no reduction with fluorouracil cream after a median follow-up time of 2.8 years.

Other agents — Oral capecitabine has been examined for the treatment and prevention of precancerous and cancerous skin lesions, particularly SCCs. A systematic review of studies documented reduction of BCCs in one small case series [162]. However, adverse effects were common, including fatigue (39 percent), diarrhea (26 percent), hand-foot syndrome (26 percent), nausea/vomiting (22 percent), anemia (13 percent), mucositis (13 percent), and warranted discontinuation in 43 percent of cases.

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Non-melanoma skin cancer (The Basics)" and "Patient education: Sunburn (The Basics)")

●Beyond the Basics topics (see "Patient education: Sunburn (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Definition and epidemiology – Basal cell carcinoma (BCC) is a common skin cancer arising from the basal layer of epidermis and its appendages. BCC is the most common malignancy in White populations and its incidence is increasing worldwide. BCC has been linked to exposure to ultraviolet (UV) light, especially during childhood. Most other risk factors act through an interaction with UV exposure. (See 'Epidemiology' above and 'Risk factors' above.)

●Clinical presentation – Approximately 80 percent of BCCs present on the face and head. The most common presentations for BCC are the nodular (picture 3G) and superficial forms (picture 3A-F), which together account for over 90 percent of cases. Although these tumors have a low metastatic potential, they are locally invasive and can be destructive of skin and the surrounding structures (picture 1A-B). (See 'Clinical presentation' above.)

●Diagnosis – Clinicians who are familiar with the clinical manifestations of BCC are often able to make the diagnosis based upon clinical examination and dermoscopic features. However, a skin biopsy is usually performed to provide pathologic confirmation of the diagnosis and determine the histologic subtype. A biopsy is particularly indicated in cases in which the diagnosis is uncertain, the patient lacks a history of BCC, the lesion exhibits features suggestive of an increased risk for tumor recurrence after treatment, or when the tumor exhibits atypical clinical features. (See 'Diagnosis' above.)

●Management – Once the diagnosis is established, appropriate treatment offers a high probability of cure, although the patient remains at increased risk for additional skin malignancies. The treatment of BCC is discussed in detail separately. (See "Treatment and prognosis of basal cell carcinoma at low risk of recurrence" and "Treatment of basal cell carcinomas at high risk for recurrence".)