- Email: [email protected]
Tattoos, inks, and cancer
Review
Tattoos, inks, and cancer Nicolas Kluger, Virve Koljonen
The introduction in the dermis of exogenous pigments and dyes to obtain a permanent design (tattooing) represents a unique in-vivo situation, where a large amount of metallic salts and organic dyes remain in the skin for the lifetime of the bearer. The potential local and systemic carcinogenic effects of tattoos and tattoo inks remain unclear. Several studies have shed light on the presence of potential carcinogenic or procarcinogenic products in tattoo inks. We extensively reviewed the literature and found 50 cases of skin cancer on tattoos: 23 cases of squamous-cell carcinoma and keratoacanthoma, 16 cases of melanoma, and 11 cases of basal-cell carcinoma. The number of skin cancers arising in tattoos is seemingly low, and this association has to be considered thus far as coincidental.
Introduction Tattooing is defined by the introduction in the dermis of exogenous pigments and dyes to obtain a permanent design.1 It has gained tremendous popularity in the past 20 years among young people in developed countries. Prevalence of tattooed individuals in the USA is estimated to be around 25% among the 18–50 year age group,2 whereas in Europe and Australia, this proportion is closer to 10%.3 The so-called typical tattooed individual is in their 30s and has 1–3 tattoos that were created in a professional tattoo parlour, with tattoos located on the trunk or arms, covering 100–300 cm² of the body surface, and usually coloured in black and red.3 Complications related to tattooing began to be reported at the end of the 19th century. Specific concerns regard the occurrence of skin cancers due to the potential introduction of carcinogenic, procarcinogenic, and toxic compounds in the skin. We did a complete and updated overview of current knowledge of tattoo ink toxicology, and reviewed clinical data related to skin tumours arising on tattoos. We also discuss the potential effects of tattooing on development of visceral cancers.
Tattoo ink toxicology: what is known? The precise composition of tattoo inks is not regulated. The US Food and Drug Administration has not approved any ink for use in tattooing, and in 2003 the European Council stressed the need for regulation of this obscure market.4 In France, recent legislation mandated that manufacturers provide the precise composition of inks, and in Germany, the use of azo pigments that can be cleaved to hazardous aromatic amines is illegal.5 Control and regulation remain difficult because most inks are purchased abroad via the internet or during tattoo conventions. Tattoo colourants include pigments, which can be inorganic metallic salts or different types of organic molecules, and organic dyes.4 Auxiliary ingredients are usually added to modify the properties of the ink and make it suitable for the tattooing procedure. The composition of inks has changed over the past 20 years. Studies were done in the 1980s and 1990s of tattoos created 10–20 years before, and their results might not apply to tattoos performed nowadays.4 Inks are less likely than before to contain inorganic salts such as mercury, cadmium, and cobalt. Aluminium, oxygen, titanium, and www.thelancet.com/oncology Vol 13 April 2012
carbon are common ingredients of tattoo inks irrespective of the colour.6,7 Several studies have shown that different colours can share a component, whereas a similar colour might not, and that some metallic salts are still commonly used.6,7 Known allergenic metals such as chromium, nickel, and cobalt have been found to exceed safe allergological limits.7 Surveyed tattoo inks were composed of several metallic salts, some found at high concentrations and others at low or trace concentrations.7 The traditional classification, based on the rule that one colour equals one metallic salt (ie, red=mercury, blue=cobalt, green=chromium, purple=magnesium, etc), which is still mentioned in some textbooks and reviews, is too restrictive and definitely out-of-date. European market studies have shown that most of the current commercially available tattoo compounds are azo pigments or polycyclic compounds, classified by their chemical constitution.4 Among the organic colourants identified in tattoo inks, 60% were azo compounds, some of which are classified as carcinogenic (anisidine, nitro-otoluidine, chloro-o-toluidine, and 3,3´-dichlorobenzidine).4 3,3´-dichlorobenzine is thought to be released from the azo pigments found in these products.4 To our knowledge, tattoo colourants or pigments have not been assessed by the International Agency for Research on Cancer as tattoo colourants, although some components of tattoo inks have been classified. Mercury, cobalt sulphate, other soluble cobalt salts, and carbon black are classed as Group 2B (possibly carcinogenic to humans), whereas cadmium and cadmium compounds are in Group 1 (carcinogenic to humans). Byproducts, some of which are absent from the initial ink, can appear after ultraviolet or laser exposures. Some of these byproducts are known to be carcinogenic to animals. According to the Toxicology Data Network’s (TOXNET) database, some components are therefore reasonably anticipated to be human carcinogens, such as 4-chlorotoluidine, or have unknown relevance to humans, such as 5-nitrotoluidine.4 Knowledge of tattoo ink toxicology has improved thanks to several groups who have studied the potential health risk of tattoo inks by analysing the composition and byproducts of tattoo ink in various circumstances in vitro.5,8–20 Experimental conditions usually include photochemical decomposition after sun, ultraviolet, or
Lancet Oncol 2012; 13: e161–68 Departments of Dermatology, Allergology, and Venereology, Institute of Clinical Medicine, University of Helsinki, Skin and Allergies Hospital (N Kluger MD), and Department of Plastic Surgery, University of Helsinki, Töölö Hospital (V Koljonen MD), Helsinki University Central Hospital, Helsinki, Finland Correspondence to: Dr Nicolas Kluger, Department of Dermatology, Skin and Allergies Hospital, Meilahdentie 2, PO Box 160, 00029 HUS, Helsinki, Finland [email protected]
For more on the TOXNET database see http://toxnet.nlm. nih.gov/cgi-bin/sis/ htmlgen?HSDB
e161
Review
Age (years) Duration of at diagnosis tattoo at presentation and sex Sharlit (1938)21
9, M
3 months
Delay of onset after tattooing/prior to diagnosis*
Tattoo location
Colour within affected area
Chronic sun exposure
Other skin cancer
NA/NA
Forehead
Indelible pencil puncture NA
NA
Allen (1967)22
NA
NA
NA
Arm
NA
NA
NA
Kirsh (1969)23† Kirsh (1972)24†
52, M
27 years
26 years/1 year
Right arm
Blue
NA
No
Wolfort et al (1974)25
55, M
29 years
24 years/5 years
Right arm
Red
NA
NA
Bartal et al (1980)26
52, F
2 years
2 years/NA
Breast (radiotherapy Black (Indian ink) field marking tattoo)
NA
No
Bartal et al (1980)26
34, NA
6 years
6 years/NA Two metastatic malignant nodules from a melanoma of the left forearm localisation
Back (radiotherapy Black (Indian Ink) field marking tattoo)
NA
No
Lee and Craig (1984)27
44, M
>20 years
>16 years/4 years
Chest
Blue
NA
No
Kircik et al (1993)28
36, F
10 years
9 years/1 year
Right scapula
Dark blue-green
NA
NA
Soroush et al (1997)29
47, M
20 years
16 years/4 years
Abdomen
Black
Outdoor recreational and occupational sun exposure, no sunburn
No
Khan et al (1999)30
44, M
25 years
NA/NA
Right arm
Black, blue
Outdoor recreational and occupational sun exposure, no sunburn
No
Stinco et al (2003)31
26, M
NA
Pre-existing naevus, modification for 3 years
Left scapula
··
NA
NA
Paradisi et al (2006)32
36, M
10 years
Prexisting naevus, modification for 1 year
Left scapula
Dark blue
Outdoor recreational and occupational sun exposure, no sunburn
No
Shariff et al (2006)33
48, M
30 years
30 years/3 months
Right arm
NA
NA
NA
Singh et al (2007)34
56, M
NA
NA/NA, recent change in colour
Right arm
NA
NA
NA
Kluger et al (2008)35
70, M
>40 years
NA/NA
Left arm
Black, dark blue
Farmer with chronic sun exposure
Developed a second melanoma on tattoo-free area
Varga et al (2011)72
28, M
5 years
4 years/pre-existing naevus, modification for 1 year after a trauma
Right arm
Black
NA
No
Kaskel et al (2000)73 also mentioned a case of melanoma on the arm over a 10-year-old tattoo without other detail. M=male. NA=data not available specifically. F=female. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Double publication of the same case.
Table 1: Main characteristics of reported cases of melanoma arising in tattoos
laser exposure of selected pigments (pigment yellow 74, pigment red 22, or pigment red 9).9–11 Several byproducts have been indentified; some have not been proven to be harmful to humans,10 whereas others such as 2-methyl5-nitroaniline (2,5-MNA), 4-nitrotoluene (4-NT), 2,5-dichloroaniline (2,5-DCA), and 1,4-dichlorobenzene (1,4-DCB)8,11 could have toxic effects in human cells or in animals.12,14 However, such experimental conditions and choice of pigments makes direct interpretation to humans difficult, particularly since the toxicities occurred in a different setting than tattooing. Some byproducts have also been detected after laser exposure, which are either absent or in lower concentrations than before exposure, shedding light on the potential carcinogenic consequences of laser tattoo removal.8 The skin is an important organ, serving as a barrier to the environment and having the capacity to metabolise chemicals that are absorbed.10 It is unclear whether the skin is able to metabolise tattoo ink byproducts. Engel and colleagues15 estimated that the mean concentration in the e162
dermis of pigments tattooed on human and pig skin was 2·53 mg/cm² (range 0·60–9·42). This result, combined with knowledge of the mean size of tattoos,3 allows estimation of a mean concentration of any introduced tattoo ink product. According to Engel and colleagues,15 about 253 mg of azo pigment red 22 (PR 22) are deposited in the dermis for a typical tattoo covering a skin area of 100 cm². However, this mean value was obtained from various tattooing circumstances. Expert tattooists achieved the lowest concentrations (0·6 mg/cm²), compared with researchers applying the experimental tattoos.15 Also, the concentration was obtained just after tattooing, without taking into account ink clearance. Using the SKH-1 tattooed mouse model,16 the same team of researchers showed natural skin clearance of PR 22 of around 32% of the initial value, and clearance after ultraviolet and laser exposure of around 60% and 51% of the mean value, respectively.17 Hazardous aromatic amines (2,5-MNA and 4-NT) were also detected.17 High performance liquid chromatography and mass spectrometry analysis of www.thelancet.com/oncology Vol 13 April 2012
Review
Age (years) at diagnosis and sex
Duration of Delay of onset after tattooing/prior to tattoo at presentation diagnosis*
Tattoo location
Colour within affected Chronic sun exposure area
Other skin cancer
NA
Bashir (1976)36
60, F
20 years
17 years/3 years
Temple (therapeutic tattoo)†
Dark pigment (carbon)
NA, but most likely (southern Iraq)
Bashir (1976)36
52, M
15 years
11 years/4 years
Temple (therapeutic tattoo)†
Dark pigment (carbon)
NA, but most likely (southern Iraq)
NA
Earley (1983)37
76, M
57·5 years
55 years/2·5 years
Shoulder
Red
NA
NA
Earley (1983)37
64, M
40 years
40 years/NA
Hand
Dark black pigment
NA, most likely
NA
Wiener and Scher (1987)38
64, M
46 years
NA/NA
Left arm
Blue-green
Yes (merchant seaman), but no other lesion related to sun exposure
No
Doumat et al (2004)39
35, M
1 year
1 year/NA
Left scapula
NA
No (applied sunblock regularly)
No
Birnie et al (2006)40
28, F
6 years
6 years/NA
Central back
Black
No
No
Kluger et al (2008)35
40, M
7 years
2–3 years/4–5 years
Shoulder and back
Blue
Yes
Several BCCs and one melanoma
Lee et al (2009)41
60, F
5 years
Eyebrow (cosmetic tattoo)
Black
NA‡
No
Omidian and Emad-Mostofi (2009)42
64, F
30 years
29 years/14 months
Upper lip§
Blue-black
NA‡
No
Omidian and Emad-Mostofi (2009)42
72, F
NA
NA/1 year
Upper lip§
Blue-black
NA‡
No
3 years/2 years
F=female. NA=data not available specifically. M=male. BCC=basel-cell carcinoma. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Applied as part of traditional medicine for treatment of headache. ‡Chronic sun exposure not mentioned, but most likely according to the precise location of tattoos. §Traditional religious Iranian tattoos.
Table 2: Main characteristics of reported cases of basal-cell carcinomas arising in tattoos
19 commercial black tattoo inks showed the presence of polycyclic aromatic hydrocarbons (PAHs) at concentrations from 0·14–201 μg/g of tattoo ink, and phenols at 0·2–385 μg/g.18 Some of the PAHs detected are possibly or probably carcinogenic or mutagenic, and generate singlet oxygen under ultraviolet irradiation. So far, it is unknown whether PAHs in black tattoo inks contribute to carcinogenic risk for individuals with tattoos. Lehner and colleagues5 found other products in commercial black inks, such as dibutylphthalate, hexachloro-1,3-butadiene, metheneamine, dibenzofuran, benzophenone, and 9-fluorenone. Some of these products might be genotoxic, cytotoxic, or generate reactive oxygen species under light exposure. Recently, Høgsberg and colleagues19 noted the presence of nanoparticles in tattoo inks; the relevance to health issues remains to be assessed. Most of these investigations of tattoo ink composition were in-vitro studies, such as chromatographic detection of compounds and analysis of cell behaviour after direct exposure to inks. The in-vivo biological relevance is far from being clear—ie, whether some of the substances detected can cause health problems when tattooed in the skin.5 The capacity of the skin to clear toxic products is not well understood. The consequences of potential byproducts produced after laser tattoo removal are uncertain, and so far, epidemiological clinical data does not support an alarming increase in skin cancers.
Skin cancers in tattoos In the past 40 years, roughly 50 cases of malignant melanoma,21–35 basal-cell carcinoma (BCC),35–42 and www.thelancet.com/oncology Vol 13 April 2012
squamous-cell carcinoma (SCC)43–48 associated with tattoos have been reported. There are also anecdotal cases of rare cutaneous malignancies, for which a true link between tattooing and occurrence of the lesion is highly speculative; these include two cases of dermatofibrosarcoma protuberans (Darier-Ferrand)49,50 that occurred 1 and 2 years after tattooing, and a leiomyosarcoma51 that occurred 9 years after tattooing. One case of B-cell lymphoma was reported in a patient with a long history of pseudolymphoma (a benign lymphocytic reaction infiltration52) on tattoos on both arms—lymphoma developed on tattooed and non-tattooed areas, most likely as a clonal response to the chronic antigenic stimulation to one component of the tattoo ink.52 Two additional reported cases of cutaneous lymphoma53,54 have been reconsidered as pseudolymphoma. There has never been a case of multiple malignant melanomas, BCCs, or SCCs occurring on a single tattoo. Additional issues are raised by the occurrence of possibly malignant lesions, such as pseudoepitheliomatous hyperplasia55–62 and keratoacanthomas.63–70 In the past few years, there has been an increase in reported cases of isolated and eruptive keratoacanthomas and pseudoepitheliomatous hyperplasia on tattoos. The reasons for this increase are unclear; does it reflect a true increase in incidence, were these complications just better recognised, or is it simply publishing bias of a trendy complication?71 Keratoacanthomas are borderline lesions whose true malignancy is a matter of debate between dermatologists and pathologists; however, some researchers consider all keratoacanthomas as SCCs.50 e163
Review
Age (years) at diagnosis and sex MacQuarrie (1966)43
NA
Cipollaro (1973)63
24, M
Goldstein (1979)64
NA
Goldstein (1979)64 Pitarch et al (2007)44
Duration of Delay of onset after Diagnosis (KA or SCC) tattooing/prior to tattoo at presentation diagnosis* 21 years
SCC
Colour within affected area
Chronic sun exposure
Other skin cancer
NA
Breast
Red
NA
4 months/2 months KA†
Arm
Red
NA
NA
NA
NA
KA
NA
Red
NA
NA
NA
NA
NA
KA
NA
Red
NA
NA
35, M
10 years
10 years/4 months
Well differentiated SCC
Right arm
Black
NA
No
Pitarch et al (2007)44
30, F
10 months
NA/NA (“recent”) NA/NA
Well differentiated SCC KA
NA
Black Red and black
NA
No
Kleinerman et al (2007)65
43, F
9 years
9 years/1 month
KA
Right calf
Red
NA, African American No origin
Chorny et al (2007)67
56, M
3 weeks
3 weeks/NA, patient Eruptive KA (x4) consult occurred soon after
Left forearm
Red, black, and yellow
NA
2 BCCs
Tan-billet et al (2007)45
51, M
4 weeks
4 weeks/NA
Right forearm
NA
Yes, several actinic keratoses
No
Kluger et al (2008)66
41, F
1·5 months
Goldenberg et al (2008)68
38, M
Ortiz and Yamauchi (2009)46
47, F
Fraga and Prossick (2010)70
6 months
NA
Tattoo location
SCC with verrucous features
6 weeks/1 month
KA
Scapula
Red
NA
NA
1 month/NA
Eruptive KA (x5)
Left forearm
NA, many colours NA
No
7–10 days
7–10 days/NA
SCC (KA-like feature)
Upper lip (cosmetic tattoo)
Red (permanent make-up ink)
No
No
54, M 55 56
NA
NA/4 weeks NA/NA
KA KA KA (x2)
Right leg Arm Right shin
Red and yellow Red Red
NA, but most likely
Actinic keratoses and melanoma in situ of the scalp
Fraga and Prossick (2010)70
62, F
NA
NA
KA
Ankle
Red
NA
NA
Fraga and Prossick (2010)70
66, M
NA
NA
KA
Leg
Red
NA
NA
Fraga and Prossick (2010)70
36, F
NA
NA
KA
Ankle
Red
NA
NA
Fraga and Prossick (2010)70
53, M
NA
NA
KA and SCC
Arm
Red
NA
NA
Fraga and Prossick (2010)70
52, M
NA
NA
KA
Arm
Red
NA
NA
Fraga and Prossick (2010)70
50, M
NA
NA
KA
Arm
Red
NA
NA
Fraga and Prossick (2010)70
49, M
NA
NA
KA
Arm
Red
NA
NA
Vitiello et al (2010)48
39, M
3 months
1 month/2 months
Eruptive KA (x8)
Left leg
Red
NA, African American NA origin
Gon Ados et al (2009)69
60, F
4 month
3 months/1 month
KA
Right leg
Red
NA
NA
Sarma et al (2010)47
79, M
NA/NA
Poorly differentiated SCC
Left forearm
Black
Yes, homeless and sun-damaged dermis
No
1 months
>50 years
KA=keratoacanthoma. SCC=squamous-cell carcinoma. NA=data not available specifically. M=male. F=female. BCC=basal-cell carcinoma. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Diagnosis made by shaving, no full thickness excision (pseudoepitheliomatous process highly suggestive of KA).
Table 3: Main characteristics of reported cases of squamous-cell carcinomas and keratoacanthomas arising in tattoos
Several cases of SCCs have been reported on tattoos,43–48 and Fraga and Prossick70 reported the largest series of 11 keratoacanthomas in tattoos. Clinicians should be careful not to classify all keratoacanthomas as SCCs. Two different entities should be considered: genuine, trauma-induced keratoacanthomas that occur rapidly after a tattooing procedure, usually within a week to a year, and classic SCC that usually develops years after tattooing. Indeed, the delay after procedure is crucial data; most of the reported cases of keratoacanthoma on a tattoo occurred within a year after tattooing.70,71 In only one case, the diagnosis of keratoacanthoma could be doubtful since the lesion occurred 7–8 years after tattooing.65 Pseudoepitheliomatous hyperplasia is also a e164
benign lesion that develops rapidly after tattooing;55–62 however, it usually presents as a verrucous lesion that can be difficult to distinguish from keratoacanthoma and SCC, clinically and pathologically.71 A full excision of these lesions (keratoacanthoma or pseudoepitheliomatous hyperplasia) is necessary to make the correct diagnosis. The characteristics of patients with a reported tumour arising in a tattoos are summarised in tables 1–5. Additional issues regarding the risk of skin cancers on tattoos include the tattooing procedure itself, and the tattoo masking the underlying skin. Indeed, tattooing on a naevus can traumatise it and trigger sudden clinical and pathological modifications prompting excision and microscopic analysis to rule out malignancy. Multiple www.thelancet.com/oncology Vol 13 April 2012
Review
tattoos can make clinical surveillance of naevi difficult. The ugly duckling sign (a mole that appears different from all others on the body) might become less visible when moles are surrounded by tattoos, and this situation is further complicated when a patient has a past history of melanoma or atypical mole syndrome. Dermoscopy is difficult since tattoo pigments appear as multiple blackblue globules, which can prevent accurate diagnosis or surveillance of lesions. All of these issues increase the risk of delayed diagnosis of a cutaneous malignancy.1
Schmidt and Christensen (1978)57§
Potential links between tattoos and skin cancers If there is a true link between tattooing and skin cancer, carcinogenesis is most likely to be the result of a multifactorial process. The trauma induced by the procedure (puncturing the skin) has typically been identified as one of the main causes. Tattoo pigments do not remain inert in the dermis; an inflammatory reaction occurs over the lifetime of the bearer, in an attempt to degrade all foreign material.74 Trauma, scars, or cutaneous chronic inflammation can underlie the development of BCCs, SCCs, and keratoacanthomas.35 Trauma specifically seems to be a crucial trigger for keratoacanthomas, since most cases occur within the first year after tattooing (table 4), whereas melanoma has not been proven to be directly associated with trauma.73 Also, the scarring process made by a tattoo is very different from other carcinogenic scars, such as a burn scar or chronic inflammation (eg, chronic hidradenitis). It is impossible to draw a direct conclusion from the carcinogenic process leading from a burn scar to SCC, and the potential carcinogenic effect of a tattoo.35 Chronic ultraviolet sun exposure has been suggested as another potential factor for skin carcinogenesis in tattoos. Tattoos might be located on ultraviolet-exposed or intermittently exposed body areas.36,39,40,42,44,45 Wiener and Scher38 suggested that pigments might alter ultraviolet absorption in the skin. Moreover, it could be postulated that a tattooed individual would show off their tattoos, especially during summer time, exposing the tattooed skin to sun and ultraviolet radiation with the potential long-term risks. However, sun habits and underlying skin condition are often not mentioned in reports, or are not accurately reported by patients. Among the reported cases of BCCs in tattoos, Wiener and Scher38 did not find any other lesion related to chronic sun exposure in a patient who was a merchant seaman; another patient applied regularly sunblock,39 and the youngest patient reported so far had no specific risk factor for non-melanoma skin cancer and a tattoo located on a non-sun-exposed location.40 Individuals with tattoos are usually aware that chronic sun exposure will alter the tattoo and cause colours to fade. Heavily tattooed individuals are reported to avoid sun exposure to protect their tattoos.74 Unfortunately, so far there has been no specific study on the sun habits of tattooed individuals. www.thelancet.com/oncology Vol 13 April 2012
Goldstein (1967)55†
Duration of Delay of onset after Tattoo Age tattooing/prior to location (years) at tattoo at diagnosis presentation diagnosis* and sex
Colour Chronic Other skin within sun affected exposure cancer area
NA
NA
NA
NA
Red
NA
NA
9 years
9 years/NA
Arm
Red
NA
NA
NA
NA
2–4 weeks/NA
NA
Red
NA
NA
Balfour et al (2003)58
27, F
1 year
2 months/10 months NA
Purple
NA
NA
Cui et al (2007)59
59, F
2 years
1 week/2 years
Leg
Red
NA
NA
Kluger et al (2008)60
30, F
7 months
1 month/6 months
Ankle
Red
NA
NA
Kluger et al (2008)60
32, F
NA
4 days/2 months
Upper back
Red
NA
NA
Kluger et al (2008)60
54, F
2·25 years
3 months/2 years
Ankle
Black
NA
NA
Then et al (2009)61
24, M
NA
1 year/NA
Leg
Red
NA
No
Biwas (2011)62
38, F
7 months
3 weeks/6 months
Ankle
Red
NA
NA
Biro and Klein 26, M (1967)56‡
NA=not available. M=male. F=female. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Author reported two cases from 15 individuals presenting with pseudoepitheliomatous reaction in a red tattoo. ‡Reaction occurred 6 weeks after the tattoo was lacerated by an accident at work. §Report of seven Danish prisoners who developed ulcerated and itchy skin tumours on red areas of tattoos, with pseudoepitheliomatous hyperplasia features overlying a granulomatous reaction.
Table 4: Main characteristics of reported cases of pseudoepitheliomatous hyperplasia arising in tattoos
Does the colour of the tattoo matter? Our review of case reports showed that most of the reported cases of melanoma and BCCs occurred on black, dark blue, or dark coloured tattoos (tables 1 and 2), whereas SCCs, keratoacanthomas, and benign pseudoepitheliomatous hyperplasia developed mainly within red tattoos (tables 3 and 4). Red and black have always been the most commonly used colours for tattoos.3 The high prevalence of such reactions in these colours might simply reflect the frequency of their use. Additionally, skin surveillance for malignancies is more difficult when dark tattoos are present. It is possible that suspicious lesions are detected and excised more rapidly when they are noted on a light coloured tattoo than on a darker tattoo. But such a statement does not explain the detection of melanoma and BCC on dark colours versus SCC and keratoacanthomas on red-coloured tattoos. It is possible that the PAHs detected in black inks have a role in the carcinogenesis of lesions. However, red is known to be the main colour responsible for hypersensitivity tattoo reaction,1 and according to our review, was the most common colour for occurrence of SCCs, keratoacanthomas, and pseudoepitheliomatous hyperplasia. Different compositions of inks might therefore have crucial roles for different pathogenic processes. Unfortunately, most studies have focused on black and e165
Review
Age (years) at diagnosis and sex
Duration of Delay of onset after tattooing/prior to tattoo at presentation diagnosis*
Sangeza et al (1992)52†
54, M
>20 years
Baker et al (2005)49
35, M
8 years
1 year/7 years
Reddy et al (2011)50‡
52, M
7 years†
West et al (2009)51
41, M
10 years
Diagnosis
Longstanding history Cutaneous B-cell of pseudolymphoma lymphoma reaction to tattoos
Tattoo location
Chronic Other skin Colour cancer sun within affected exposure area
Arm
Red
NA
Lymphoma also developed in tattoo-free area
Dermatofibrosarcoma Left forearm Black protuberans
NA
No
2 years/5 years
Dermatofibrosarcoma Right upper protuberans back
Red
NA
NA
9 years/1 year
Leiomyosarcoma
Black
NA
No
Right arm
M=male. NA=data not available specifically. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Two previous cases of lymphoma on a tattoo have been reclassified as pseudolymphoma (a reactive benign polyclonal lymphocyte infiltration of the skin)53,54 ‡Patient had a previous tattoo 15 years before presentation on the same location, which was partially removed by laser 5 years later. A new tattoo was designed over the same site 7 years before presentation (ie, 2 years after laser removal of the original tattoo).
Table 5: Main characteristics of reported cases of miscellaneous tumours arising in tattoos
red inks. To our knowledge, except for the presence of metallic salts,6,7 the presence of potential carcinogenic components in other colours has not been investigated. Finally, there might be other undetermined factors, such as a genetic predisposition and viruses in the skin, with potential role in carcinogenesis associated with tattoos. Overall, the clinical relevance of malignancies in tattoos is unclear. In-vivo data are needed, such as analysis of ink composition at the site of cancer development compared with the composition on an unaffected area of the tattoo that contains the same colour. Compared with older case reports, it seems that in more recent reports patients who develop skin cancers on tattoos are younger and have a shorter delay since tattoo application.35 This observation could have two explanations. Skin cancer in younger people is unusual, but not exceptional nowadays. The younger age of patients with reported malignancies on tattoos might simply reflect this tendency. As tattooing becomes more popular, the likelihood of having a coincidental malignant lesion on a tattoo increases, and the risk is higher the more tattoos the individual has.35 Alternately, it could be speculated that the composition of ink is changing. Despite the presence of metallic salts such as mercury, getting a tattoo was possibly safer in the past than it is now (from a toxicological point of view), and new compounds with potential carcinogenic properties in recent inks might be responsible for the development of skin cancers.
Tattoo: a potential risk factor for inner organ cancers? Tattoo pigments reside mainly in the dermis, either within fibroblasts or between collagen bundles.74 Macrophages, in an attempt to eliminate the pigments, assimilate them and migrate to the lymphatic vessels then to the regional lymph nodes. Cases of macroscopic, often asymptomatic, lymph nodes filled with black pigments have been reported.75 As a generalisation, any individual with a tattoo might have a e166
regional black lymph node. Black lymph nodes can also become palpable, raising the concern for potential infection or malignancy and leading to lymph node excision75 or dissection by a surgeon who suspects metastatic melanoma. A diagnosis of metastastic melanoma might be difficult for a pathologist to assess because of the presence of exogenous pigments.76 Overall, the potential pathological relevance of a lymph node containing tattoo pigment has never been assessed. Engel and colleagues,17 using an SKH-I mouse model, noted that pigment concentration of PR 22 that was intradermally injected on mouse skin decreased by 30% within 6 weeks after tattooing. Lehner and colleagues20 extracted various red pigments (PR 9, 12, and 112) from tattooed skin of deceased patients (duration of tattoo unknown) and estimated a concentration of 0·002–0·110 mg/cm². They estimated that the remaining pigment concentration in these red tattoos was 1–13% of the original concentration.20 The precise kinetics of the decrease over time of tattoo pigments remains to be assessed on standardised human skin samples. This decrease is due to elimination of some of the pigment during the healing phase, decomposition by repeated solar radiation, and migration into the lymph nodes. So far, there is no data regarding the potential localisation of tattoo byproducts elsewhere in the body. Nevertheless, some researchers stress that the long contact time between skin and tattoos allows continued exposure to toxic or allergenic metals with consequent accumulation in the body.7 The potential diffusion of nanoparticles also remains speculative.19 Tattoo pigments might be discharged from the body without any further health-related complications.
Conclusions With roughly 50 case reports of cutaneous malignancies on tattoos over the past 40 years, and millions of tattooed individuals worldwide, the number of cases remains at the level of background noise. So far, this association has to be considered as coincidental. www.thelancet.com/oncology Vol 13 April 2012
Review
Search strategy and selection criteria References for this Review were found through a search of PubMed by use of the terms “tattoo”, “tattoos”, or “tattooing”. Since 2006, NK has collected articles on health-related issues associated with tattooing, by performing a prospective weekly search of PubMed using the above keywords. An additional search was done using the terms “tattoo or tattoos or tattooing” AND “cancer”, “skin cancer”, “melanoma”, “squamous-cell carcinoma”, “basal-cell carcinoma”, “keratoacanthoma”, “pseudo-epitheliomatous hyperplasia”, “lymphoma”, or “sarcoma”. Reference lists of relevant articles were reviewed. One conference abstract was included because it related directly to the topic. No date restrictions were used. Only articles published in English and French were included. Selected references on associated topics were also included.
6 7 8 9
10
11
12
13
14
Unfortunately, so far only a few groups are investigating the association between tattoos and skin tumours, and large-scale studies of clinical and epidemiological factors are lacking. Additional, in-vivo investigations of cutaneous concentrations of pigments and byproducts are needed. Large cohort follow-up of tattooed individuals would help to assess whether or not tattooing is an independent risk factor for melanoma, non-melanoma skin cancer, and visceral cancers. The addition of tattoo data in national skin cancer registries would help define the true prevalence of malignancies in tattoos. There are no data available regarding the sun habits of tattooed individuals, who are thought to avoid sun to protect their tattoos. Permanent tattooing with its health-related issues and potential carcinogenic effect is a topic that should not be neglected by health services, and could be an important upcoming health issue. Regulation of tattoo ink manufacturing is urgently needed. Contributors NK had the original idea for the article. Both authors contributed to conception and design of the report, and the acquisition, analysis, and interpretation of data. Both did the literature search, wrote and edited the report, and approved it for publication.
15
16
17
18
19 20
21 22 23 24 25 26
Conflicts of interest We declare that we have no conflicts of interest. References 1 Kluger N. Cutaneous complications related to permanent decorative tatooing. Expert Rev Clin Immunol 2010; 6: 363–71. 2 Laumann AE, Derick AJ. Tattoos and body piercings in the United States: a national data set. J Am Acad Dermatol 2006; 55: 413–21. 3 Klügl I, Hiller KA, Landthaler M, Bäumler W. Incidence of health problems associated with tattooed skin: a nation-wide survey in German-speaking countries. Dermatology 2010; 221: 43–50. 4 Bäumler W, Vasold R, Lundsgaard J, Talberg HJ. Chemical used in tattooing and permanent make up products. In: Papameletiou D, Schwela D, Zenie A, eds. Workshop on Technical/Scientific and Regulatory Issues on the Safety of Tattoos, Body Piercing and of Related Practices. Ispra, VA: European Commission, 2003: 21–36. 5 Lehner K, Santarelli F, Vasold R, König B, Landthaler M, Bäumler W. Black tattoo inks are a source of problematic substances such as dibutyl phthalate. Contact Dermatitis 2011; 65: 231–38.
www.thelancet.com/oncology Vol 13 April 2012
27 28 29
30 31 32
33
Beute TC, Miller CH, Timko AL, Ross EV. In vitro spectral analysis of tattoo pigments. Dermatol Surg 2008; 34: 508–15. Forte G, Petrucci F, Cristaudo A, Bocca B. Market survey on toxic metals contained in tattoo inks. Sci Total Environ 2009; 407: 5997–6002. Vasold R, Engel E, König B, Landthaler M, Bäumler W. Health risks of tattoo colors. Anal Bioanal Chem 2008; 391: 9–13. Cui Y, Spann AP, Couch LH, et al. Photodecomposition of Pigment Yellow 74, a pigment used in tattoo inks. Photochem Photobiol 2004; 80: 175–84. Cui Y, Churchwell MI, Couch LH, Doerge DR, Howard PC. Metabolism of pigment yellow 74 by rat and human microsomal proteins. Drug Metab Dispos 2005; 33: 1459–65. Vasold R, Naarmann N, Ulrich H, et al. Tattoo pigments are cleaved by laser light-the chemical analysis in vitro provide evidence for hazardous compounds. Photochem Photobiol 2004; 80: 185–90. Engel E, Santarelli F, Vasold R, et al. Establishment of an extraction method for the recovery of tattoo pigments from human skin using HPLC diode array detector technology. Anal Chem 2006; 78: 6440–47. Engel E, Spannberger A, Vasold R, König B, Landthaler M, Bäumler W. Photochemical cleavage of a tattoo pigment by UVB radiation or natural sunlight. J Dtsch Dermatol Ges 2007; 5: 583–89. Wang L, Yan J, Hardy W, Mosley C, Wang S, Yu H. Light-induced mutagenicity in Salmonella TA102 and genotoxicity/cytotoxicity in human T-cells by 3,3´-dichlorobenzidine: a chemical used in the manufacture of dyes and pigments and in tattoo inks. Toxicology 2005; 207: 411–18. Engel E, Santarelli F, Vasold R, et al. Modern tattoos cause high concentrations of hazardous pigments in skin. Contact Dermatitis 2008; 58: 228–33. Gopee NV, Cui Y, Olson G, et al. Response of mouse skin to tattooing: use of SKH-1 mice as a surrogate modelfor human tattooing. Toxicol Appl Pharmacol 2005; 209: 145–58. Engel E, Vasold R, Santarelli F, et al. Tattooing of skin results in transportation and light-induced decomposition of tattoo pigments—a first quantification in vivo using a mouse model. Exp Dermatol 2010; 19: 54–60. Regensburger J, Lehner K, Maisch T, et al. Tattoo inks contain polycyclic aromatic hydrocarbons that additionally generate deleterious singlet oxygen. Exp Dermatol 2010; 19: 275–81. Høgsberg T, Loeschner K, Löf D, Serup J. Tattoo inks of general usage contain nanoparticles. Br J Dermatol 2011; 165: 1210–18. Lehner K, Santarelli F, Penning R, et al. The decrease of pigment concentration in red tattooed skin years after tattooing. J Eur Acad Dermatol Venereol 2011; 25: 1340–45. Sharlit H. Melanoma caused by indelible pencil. Arch Dermatol 1938; 37: 301–06. Allen AC. The skin. St Louis: CV Mosby Co, 1954. Kirsch N. Malignant melanoma developing in a tattoo. Arch Dermatol 1969; 99: 596–98. Kirsch N. Malignant melanoma developing in a tattoo. Int J Dermatol 1972; 11: 16–20. Wolfort FC, Hoopes JE, Filtze HS, Cochran TC. Superficial melanoma in a tattoo. Br J Plast Surg 1974; 27: 303–04. Bartal AH, Cohen Y, Robinson E. Malignant melanoma arising at tattoo sites used for radiotherapy field marking. Br J Radiol 1980; 3: 913–14. Lee YT, Craig JR. Melanoma in a tattoo of the breast. J Surg Oncol 1984; 25: 100–01. Kircik L, Armus S, van den Broek H. Malignant melanoma in a tattoo. Int J Dermatol 1993; 32: 297–98. Soroush V, Gurevitch AW, Peng SK. Malignant melanoma in a tattoo: case report and review of the literature. Cutis 1997; 59: 111–12. Khan IU, Moiemen NS, Firth J, Frame JD. Malignant melanoma disguised by a tattoo. Br J Plast Surg 1999; 52: 598. Stinco G, De Francesco V, Frattasio A, Quinkenstein E, Patrone P. Malignant melanoma in a tattoo. Dermatology 2003; 206: 345–46. Paradisi A, Capizzi R, De Simone C, Fossati B, Proietti I, Amerio PL. Malignant melanoma in a tattoo: case report and review of the literature. Melanoma Res 2006; 16: 375–76. Shariff Z, Tehrani H, Jagadeesan J, Hardwicke J. Artwork: to be studied. Dermatol Online J 2006; 12: 21.
e167
Review
34
35
36 37 38 39 40 41 42 43 44
45 46
47
48
49 50
51
52
53 54
55
56
e168
Singh RS, Hafeez Diwan A, Prieto VG. Potential diagnostic pitfalls in melanoma arising in a cutaneous tattoo. Histopathology 2007; 51: 283–85. Kluger N, Phan A, Debarbieux S, Balme L, Thomas L. Skin cancers arising in tattoos: coincidental or not? Dermatology 2008; 217: 219–21. Bashir AH. Basal cell carcinoma in tattoos: report of two cases. Br J Plast Surg 1976; 29: 288–90. Earley MJ. Basal cell carcinoma arising in tattoos: a clinical report of two cases. Br J Plast Surg 1983; 36: 258–59. Wiener DA, Scher RK. Basal cell carcinoma arising in a tattoo. Cutis 1987; 39: 125–26. Doumat F, Kaise W, Barbaud A, Schmutz JL. Basal cell carcinoma in a tattoo. Dermatology 2004; 208: 181–82. Birnie AJ, Kulkarni K, Varma S. Basal cell carcinoma arising in a tattoo. Clin Exp Dermatol 2006; 31: 820–21. Lee JS, Park J, Kim SM, Yun SK, Kim HU. Basal cell carcinoma arising in a tattooed eyebrow. Ann Dermatol 2009; 21: 281–84. Omidian M, Emad-Mostofi N. Basal cell carcinoma arising from traditional tattoo. Arch Iranian Med 2009; 12: 198. MacQuarrie DG. An unusual breast cancer: squamous cell carcinoma arising in a tattoo. Minn Med 1966; 49: 799–801. Pitarch G, Martínez-Menchón T, Martínez-Aparicio A, Sánchez-Carazo JL, Muñoz D, Fortea JM. Squamous cell carcinoma over tattoos. J Am Acad Dermatol 2007; 56: 1072–73. Tan-Billet J, Khan U, Roberson D, Rao B. Squamous cell carcinoma developing in a tattoo. J Am Acad Dermatol 2007; 56 (suppl 2): 150. Ortiz A, Yamauchi PS. Rapidly growing squamous cell carcinoma from permanent makeup tattoo. J Am Acad Dermatol 2009; 60: 1073–74. Sarma DP, Dentlinger RB, Forystek AM, Stevens T, Huerter C. Poorly differentiated squamous cell carcinoma arising in tattooed skin. Case Report Med 2010; 2010: 431813. Vitiello M, Echeverria B, Romanelli P, Abuchar A, Kerdel F. Multiple eruptive keratoacanthomas arising in a tattoo. J Clin Aesthet Dermatol 2010; 3: 54–55. Baker PA, O’Dowd GJ, Khan IU. Dermatofibrosarcoma protuberans arising in a decorative tattoo. Sarcoma 2005; 9: 37–41. Reddy KK, Hanke CW, Tierney EP. Malignancy arising within cutaneous tattoos: case of dermatofibrosarcoma protuberans and review of literature. J Drugs Dermatol 2011; 10: 837–42. West CC, Morritt AN, Pedelty L, Lam DG. Cutaneous leiomyosarcoma arising in a tattoo—‘a tumour with no humour’. J Plast Reconstr Aesthet Surg 2009; 62: 79–80. Sangueza OP, Yadav S, White CR Jr, Braziel RM. Evolution of B-cell lymphoma from pseudolymphoma. A multidisciplinary approach using histology, immunohistochemistry, and Southern blot analysis. Am J Dermatopathol 1992; 14: 408–13. Auger C. Cancer sur tatouage. Laval Med 1943; 8: 300. Armiger WG, Caldwell EH. Primary lesion of a non-Hodgkin’s lymphoma occurring in a skin tatoo: case report. Plast Reconstr Surg 1978; 62: 125–27. Goldstein N. Mercury-cadmium sensitivity in tattoos. A photoallergic reaction in red pigment. Ann Intern Med 1967; 67: 984–89. Biro L, Klein WP. Unusual complications of mercurial (cinnabar) tattoo. Generalized eczematous eruption following laceration of a tattoo. Arch Dermatol 1967; 96: 165–67.
57 58
59
60
61 62 63 64 65
66
67 68
69 70
71
72
73
74
75 76
Schmidt H, Christensen HE. Red poster paint tattoo granuloma. Arch Dermatol 1978; 114: 965–66. Balfour E, Olhoffer I, Leffell D, Handerson T. Massive pseudoepitheliomatous hyperplasia: an unusual reaction to a tattoo. Am J Dermatopathol 2003; 25: 338–40. Cui W, McGregor DH, Stark SP, Ulusarac O, Mathur SC. Pseudoepitheliomatous hyperplasia—an unusual reaction following tattoo: report of a case and review of the literature. Int J Dermatol 2007; 46: 743–45. Kluger N, Durand L, Minier-Thoumin C, et al. Pseudoepitheliomatous epidermal hyperplasia in tattoos: report of three cases. Am J Clin Dermatol 2008; 9: 337–40. Then M, Mark Boustred A, Clarke LE. Keratoacanthomatous hyperplasia in response to a tattoo. Dermatol Surg 2009; 35: 685–86. Biswas A. Pseudoepitheliomatous tattoo reaction. Diagnostic Histopathology 2011; 17: 272–75. Cipollaro VA. Keratoacanthoma developing in a tattoo. Cutis 1973; 11: 809–10. Goldstein N. IV. Complications from tattoos. J Dermatol Surg Oncol 1979; 5: 869–78. Kleinerman R, Greenspan A, Hale EK. Mohs micrographic surgery for an unusual case of keratoacanthoma arising from a longstanding tattoo. J Drugs Dermatol 2007; 6: 931–32. Kluger N, Minier-Thoumin C, Plantier F. Keratoacanthoma occurring within the red dye of a tattoo. J Cutan Pathol 2008; 35: 504–07. Chorny JA, Stephens FV, Cohen JL. Eruptive keratoacanthomas in a new tattoo. Arch Dermatol 2007; 143: 1457–58. Goldenberg G, Patel S, Patel MJ, Williford P, Sanguenza O. Eruptive squamous cell carcinomas, keratoacanthoma type, arising in a multicolor tattoo. J Cutan Pathol 2008; 35: 62–64. Gon Ados S, Minelli L, Meissner MC. Keratoacanthoma in a tattoo. Dermatol Online J 2009; 15: 9. Fraga GR, Prossick TA. Tattoo-associated keratoacanthomas: a series of 8 patients with 11 keratoacanthomas. J Cutan Pathol 2010; 37: 85–90. Kluger N. Issues with keratoacanthoma, pseudoepitheliomatous hyperplasia and squamous cell carcinoma within tattoos: a clinical point of view. J Cutan Pathol 2010; 37: 812–13. Varga E, Korom I, Varga J, Kohán J, Kemény L, Oláh J. Melanoma and melanocytic nevi in decorative tattoos: three case reports. J Cutan Pathol 2011; 38: 994–98. Kaskel P, Kind P, Sander S, Peter RU, Krähn G. Trauma and melanoma formation: a true association? Br J Dermatol 2000; 143: 749–53. Kluger N, Plantier F, Moguelet P, Fraitag S. Tattoos: natural history and histopathology of cutaneous reactions. Ann Dermatol Venereol 2011; 138: 146–54. Kluger N, Cohen-Valensi R, Nezri M. Black lymph nodes—and a colourful skin. Lancet 2008; 371: 1214. Jaigirdar AA, Yeh MW, Sharifi E, Browne LW, Leong SP. Coexisting tattoo pigment and metastatic melanoma in the same sentinel lymph node. J Cutan Med Surg 2009; 13: 321–25.
www.thelancet.com/oncology Vol 13 April 2012
Tattoos, inks, and cancer Nicolas Kluger, Virve Koljonen
The introduction in the dermis of exogenous pigments and dyes to obtain a permanent design (tattooing) represents a unique in-vivo situation, where a large amount of metallic salts and organic dyes remain in the skin for the lifetime of the bearer. The potential local and systemic carcinogenic effects of tattoos and tattoo inks remain unclear. Several studies have shed light on the presence of potential carcinogenic or procarcinogenic products in tattoo inks. We extensively reviewed the literature and found 50 cases of skin cancer on tattoos: 23 cases of squamous-cell carcinoma and keratoacanthoma, 16 cases of melanoma, and 11 cases of basal-cell carcinoma. The number of skin cancers arising in tattoos is seemingly low, and this association has to be considered thus far as coincidental.
Introduction Tattooing is defined by the introduction in the dermis of exogenous pigments and dyes to obtain a permanent design.1 It has gained tremendous popularity in the past 20 years among young people in developed countries. Prevalence of tattooed individuals in the USA is estimated to be around 25% among the 18–50 year age group,2 whereas in Europe and Australia, this proportion is closer to 10%.3 The so-called typical tattooed individual is in their 30s and has 1–3 tattoos that were created in a professional tattoo parlour, with tattoos located on the trunk or arms, covering 100–300 cm² of the body surface, and usually coloured in black and red.3 Complications related to tattooing began to be reported at the end of the 19th century. Specific concerns regard the occurrence of skin cancers due to the potential introduction of carcinogenic, procarcinogenic, and toxic compounds in the skin. We did a complete and updated overview of current knowledge of tattoo ink toxicology, and reviewed clinical data related to skin tumours arising on tattoos. We also discuss the potential effects of tattooing on development of visceral cancers.
Tattoo ink toxicology: what is known? The precise composition of tattoo inks is not regulated. The US Food and Drug Administration has not approved any ink for use in tattooing, and in 2003 the European Council stressed the need for regulation of this obscure market.4 In France, recent legislation mandated that manufacturers provide the precise composition of inks, and in Germany, the use of azo pigments that can be cleaved to hazardous aromatic amines is illegal.5 Control and regulation remain difficult because most inks are purchased abroad via the internet or during tattoo conventions. Tattoo colourants include pigments, which can be inorganic metallic salts or different types of organic molecules, and organic dyes.4 Auxiliary ingredients are usually added to modify the properties of the ink and make it suitable for the tattooing procedure. The composition of inks has changed over the past 20 years. Studies were done in the 1980s and 1990s of tattoos created 10–20 years before, and their results might not apply to tattoos performed nowadays.4 Inks are less likely than before to contain inorganic salts such as mercury, cadmium, and cobalt. Aluminium, oxygen, titanium, and www.thelancet.com/oncology Vol 13 April 2012
carbon are common ingredients of tattoo inks irrespective of the colour.6,7 Several studies have shown that different colours can share a component, whereas a similar colour might not, and that some metallic salts are still commonly used.6,7 Known allergenic metals such as chromium, nickel, and cobalt have been found to exceed safe allergological limits.7 Surveyed tattoo inks were composed of several metallic salts, some found at high concentrations and others at low or trace concentrations.7 The traditional classification, based on the rule that one colour equals one metallic salt (ie, red=mercury, blue=cobalt, green=chromium, purple=magnesium, etc), which is still mentioned in some textbooks and reviews, is too restrictive and definitely out-of-date. European market studies have shown that most of the current commercially available tattoo compounds are azo pigments or polycyclic compounds, classified by their chemical constitution.4 Among the organic colourants identified in tattoo inks, 60% were azo compounds, some of which are classified as carcinogenic (anisidine, nitro-otoluidine, chloro-o-toluidine, and 3,3´-dichlorobenzidine).4 3,3´-dichlorobenzine is thought to be released from the azo pigments found in these products.4 To our knowledge, tattoo colourants or pigments have not been assessed by the International Agency for Research on Cancer as tattoo colourants, although some components of tattoo inks have been classified. Mercury, cobalt sulphate, other soluble cobalt salts, and carbon black are classed as Group 2B (possibly carcinogenic to humans), whereas cadmium and cadmium compounds are in Group 1 (carcinogenic to humans). Byproducts, some of which are absent from the initial ink, can appear after ultraviolet or laser exposures. Some of these byproducts are known to be carcinogenic to animals. According to the Toxicology Data Network’s (TOXNET) database, some components are therefore reasonably anticipated to be human carcinogens, such as 4-chlorotoluidine, or have unknown relevance to humans, such as 5-nitrotoluidine.4 Knowledge of tattoo ink toxicology has improved thanks to several groups who have studied the potential health risk of tattoo inks by analysing the composition and byproducts of tattoo ink in various circumstances in vitro.5,8–20 Experimental conditions usually include photochemical decomposition after sun, ultraviolet, or
Lancet Oncol 2012; 13: e161–68 Departments of Dermatology, Allergology, and Venereology, Institute of Clinical Medicine, University of Helsinki, Skin and Allergies Hospital (N Kluger MD), and Department of Plastic Surgery, University of Helsinki, Töölö Hospital (V Koljonen MD), Helsinki University Central Hospital, Helsinki, Finland Correspondence to: Dr Nicolas Kluger, Department of Dermatology, Skin and Allergies Hospital, Meilahdentie 2, PO Box 160, 00029 HUS, Helsinki, Finland [email protected]
For more on the TOXNET database see http://toxnet.nlm. nih.gov/cgi-bin/sis/ htmlgen?HSDB
e161
Review
Age (years) Duration of at diagnosis tattoo at presentation and sex Sharlit (1938)21
9, M
3 months
Delay of onset after tattooing/prior to diagnosis*
Tattoo location
Colour within affected area
Chronic sun exposure
Other skin cancer
NA/NA
Forehead
Indelible pencil puncture NA
NA
Allen (1967)22
NA
NA
NA
Arm
NA
NA
NA
Kirsh (1969)23† Kirsh (1972)24†
52, M
27 years
26 years/1 year
Right arm
Blue
NA
No
Wolfort et al (1974)25
55, M
29 years
24 years/5 years
Right arm
Red
NA
NA
Bartal et al (1980)26
52, F
2 years
2 years/NA
Breast (radiotherapy Black (Indian ink) field marking tattoo)
NA
No
Bartal et al (1980)26
34, NA
6 years
6 years/NA Two metastatic malignant nodules from a melanoma of the left forearm localisation
Back (radiotherapy Black (Indian Ink) field marking tattoo)
NA
No
Lee and Craig (1984)27
44, M
>20 years
>16 years/4 years
Chest
Blue
NA
No
Kircik et al (1993)28
36, F
10 years
9 years/1 year
Right scapula
Dark blue-green
NA
NA
Soroush et al (1997)29
47, M
20 years
16 years/4 years
Abdomen
Black
Outdoor recreational and occupational sun exposure, no sunburn
No
Khan et al (1999)30
44, M
25 years
NA/NA
Right arm
Black, blue
Outdoor recreational and occupational sun exposure, no sunburn
No
Stinco et al (2003)31
26, M
NA
Pre-existing naevus, modification for 3 years
Left scapula
··
NA
NA
Paradisi et al (2006)32
36, M
10 years
Prexisting naevus, modification for 1 year
Left scapula
Dark blue
Outdoor recreational and occupational sun exposure, no sunburn
No
Shariff et al (2006)33
48, M
30 years
30 years/3 months
Right arm
NA
NA
NA
Singh et al (2007)34
56, M
NA
NA/NA, recent change in colour
Right arm
NA
NA
NA
Kluger et al (2008)35
70, M
>40 years
NA/NA
Left arm
Black, dark blue
Farmer with chronic sun exposure
Developed a second melanoma on tattoo-free area
Varga et al (2011)72
28, M
5 years
4 years/pre-existing naevus, modification for 1 year after a trauma
Right arm
Black
NA
No
Kaskel et al (2000)73 also mentioned a case of melanoma on the arm over a 10-year-old tattoo without other detail. M=male. NA=data not available specifically. F=female. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Double publication of the same case.
Table 1: Main characteristics of reported cases of melanoma arising in tattoos
laser exposure of selected pigments (pigment yellow 74, pigment red 22, or pigment red 9).9–11 Several byproducts have been indentified; some have not been proven to be harmful to humans,10 whereas others such as 2-methyl5-nitroaniline (2,5-MNA), 4-nitrotoluene (4-NT), 2,5-dichloroaniline (2,5-DCA), and 1,4-dichlorobenzene (1,4-DCB)8,11 could have toxic effects in human cells or in animals.12,14 However, such experimental conditions and choice of pigments makes direct interpretation to humans difficult, particularly since the toxicities occurred in a different setting than tattooing. Some byproducts have also been detected after laser exposure, which are either absent or in lower concentrations than before exposure, shedding light on the potential carcinogenic consequences of laser tattoo removal.8 The skin is an important organ, serving as a barrier to the environment and having the capacity to metabolise chemicals that are absorbed.10 It is unclear whether the skin is able to metabolise tattoo ink byproducts. Engel and colleagues15 estimated that the mean concentration in the e162
dermis of pigments tattooed on human and pig skin was 2·53 mg/cm² (range 0·60–9·42). This result, combined with knowledge of the mean size of tattoos,3 allows estimation of a mean concentration of any introduced tattoo ink product. According to Engel and colleagues,15 about 253 mg of azo pigment red 22 (PR 22) are deposited in the dermis for a typical tattoo covering a skin area of 100 cm². However, this mean value was obtained from various tattooing circumstances. Expert tattooists achieved the lowest concentrations (0·6 mg/cm²), compared with researchers applying the experimental tattoos.15 Also, the concentration was obtained just after tattooing, without taking into account ink clearance. Using the SKH-1 tattooed mouse model,16 the same team of researchers showed natural skin clearance of PR 22 of around 32% of the initial value, and clearance after ultraviolet and laser exposure of around 60% and 51% of the mean value, respectively.17 Hazardous aromatic amines (2,5-MNA and 4-NT) were also detected.17 High performance liquid chromatography and mass spectrometry analysis of www.thelancet.com/oncology Vol 13 April 2012
Review
Age (years) at diagnosis and sex
Duration of Delay of onset after tattooing/prior to tattoo at presentation diagnosis*
Tattoo location
Colour within affected Chronic sun exposure area
Other skin cancer
NA
Bashir (1976)36
60, F
20 years
17 years/3 years
Temple (therapeutic tattoo)†
Dark pigment (carbon)
NA, but most likely (southern Iraq)
Bashir (1976)36
52, M
15 years
11 years/4 years
Temple (therapeutic tattoo)†
Dark pigment (carbon)
NA, but most likely (southern Iraq)
NA
Earley (1983)37
76, M
57·5 years
55 years/2·5 years
Shoulder
Red
NA
NA
Earley (1983)37
64, M
40 years
40 years/NA
Hand
Dark black pigment
NA, most likely
NA
Wiener and Scher (1987)38
64, M
46 years
NA/NA
Left arm
Blue-green
Yes (merchant seaman), but no other lesion related to sun exposure
No
Doumat et al (2004)39
35, M
1 year
1 year/NA
Left scapula
NA
No (applied sunblock regularly)
No
Birnie et al (2006)40
28, F
6 years
6 years/NA
Central back
Black
No
No
Kluger et al (2008)35
40, M
7 years
2–3 years/4–5 years
Shoulder and back
Blue
Yes
Several BCCs and one melanoma
Lee et al (2009)41
60, F
5 years
Eyebrow (cosmetic tattoo)
Black
NA‡
No
Omidian and Emad-Mostofi (2009)42
64, F
30 years
29 years/14 months
Upper lip§
Blue-black
NA‡
No
Omidian and Emad-Mostofi (2009)42
72, F
NA
NA/1 year
Upper lip§
Blue-black
NA‡
No
3 years/2 years
F=female. NA=data not available specifically. M=male. BCC=basel-cell carcinoma. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Applied as part of traditional medicine for treatment of headache. ‡Chronic sun exposure not mentioned, but most likely according to the precise location of tattoos. §Traditional religious Iranian tattoos.
Table 2: Main characteristics of reported cases of basal-cell carcinomas arising in tattoos
19 commercial black tattoo inks showed the presence of polycyclic aromatic hydrocarbons (PAHs) at concentrations from 0·14–201 μg/g of tattoo ink, and phenols at 0·2–385 μg/g.18 Some of the PAHs detected are possibly or probably carcinogenic or mutagenic, and generate singlet oxygen under ultraviolet irradiation. So far, it is unknown whether PAHs in black tattoo inks contribute to carcinogenic risk for individuals with tattoos. Lehner and colleagues5 found other products in commercial black inks, such as dibutylphthalate, hexachloro-1,3-butadiene, metheneamine, dibenzofuran, benzophenone, and 9-fluorenone. Some of these products might be genotoxic, cytotoxic, or generate reactive oxygen species under light exposure. Recently, Høgsberg and colleagues19 noted the presence of nanoparticles in tattoo inks; the relevance to health issues remains to be assessed. Most of these investigations of tattoo ink composition were in-vitro studies, such as chromatographic detection of compounds and analysis of cell behaviour after direct exposure to inks. The in-vivo biological relevance is far from being clear—ie, whether some of the substances detected can cause health problems when tattooed in the skin.5 The capacity of the skin to clear toxic products is not well understood. The consequences of potential byproducts produced after laser tattoo removal are uncertain, and so far, epidemiological clinical data does not support an alarming increase in skin cancers.
Skin cancers in tattoos In the past 40 years, roughly 50 cases of malignant melanoma,21–35 basal-cell carcinoma (BCC),35–42 and www.thelancet.com/oncology Vol 13 April 2012
squamous-cell carcinoma (SCC)43–48 associated with tattoos have been reported. There are also anecdotal cases of rare cutaneous malignancies, for which a true link between tattooing and occurrence of the lesion is highly speculative; these include two cases of dermatofibrosarcoma protuberans (Darier-Ferrand)49,50 that occurred 1 and 2 years after tattooing, and a leiomyosarcoma51 that occurred 9 years after tattooing. One case of B-cell lymphoma was reported in a patient with a long history of pseudolymphoma (a benign lymphocytic reaction infiltration52) on tattoos on both arms—lymphoma developed on tattooed and non-tattooed areas, most likely as a clonal response to the chronic antigenic stimulation to one component of the tattoo ink.52 Two additional reported cases of cutaneous lymphoma53,54 have been reconsidered as pseudolymphoma. There has never been a case of multiple malignant melanomas, BCCs, or SCCs occurring on a single tattoo. Additional issues are raised by the occurrence of possibly malignant lesions, such as pseudoepitheliomatous hyperplasia55–62 and keratoacanthomas.63–70 In the past few years, there has been an increase in reported cases of isolated and eruptive keratoacanthomas and pseudoepitheliomatous hyperplasia on tattoos. The reasons for this increase are unclear; does it reflect a true increase in incidence, were these complications just better recognised, or is it simply publishing bias of a trendy complication?71 Keratoacanthomas are borderline lesions whose true malignancy is a matter of debate between dermatologists and pathologists; however, some researchers consider all keratoacanthomas as SCCs.50 e163
Review
Age (years) at diagnosis and sex MacQuarrie (1966)43
NA
Cipollaro (1973)63
24, M
Goldstein (1979)64
NA
Goldstein (1979)64 Pitarch et al (2007)44
Duration of Delay of onset after Diagnosis (KA or SCC) tattooing/prior to tattoo at presentation diagnosis* 21 years
SCC
Colour within affected area
Chronic sun exposure
Other skin cancer
NA
Breast
Red
NA
4 months/2 months KA†
Arm
Red
NA
NA
NA
NA
KA
NA
Red
NA
NA
NA
NA
NA
KA
NA
Red
NA
NA
35, M
10 years
10 years/4 months
Well differentiated SCC
Right arm
Black
NA
No
Pitarch et al (2007)44
30, F
10 months
NA/NA (“recent”) NA/NA
Well differentiated SCC KA
NA
Black Red and black
NA
No
Kleinerman et al (2007)65
43, F
9 years
9 years/1 month
KA
Right calf
Red
NA, African American No origin
Chorny et al (2007)67
56, M
3 weeks
3 weeks/NA, patient Eruptive KA (x4) consult occurred soon after
Left forearm
Red, black, and yellow
NA
2 BCCs
Tan-billet et al (2007)45
51, M
4 weeks
4 weeks/NA
Right forearm
NA
Yes, several actinic keratoses
No
Kluger et al (2008)66
41, F
1·5 months
Goldenberg et al (2008)68
38, M
Ortiz and Yamauchi (2009)46
47, F
Fraga and Prossick (2010)70
6 months
NA
Tattoo location
SCC with verrucous features
6 weeks/1 month
KA
Scapula
Red
NA
NA
1 month/NA
Eruptive KA (x5)
Left forearm
NA, many colours NA
No
7–10 days
7–10 days/NA
SCC (KA-like feature)
Upper lip (cosmetic tattoo)
Red (permanent make-up ink)
No
No
54, M 55 56
NA
NA/4 weeks NA/NA
KA KA KA (x2)
Right leg Arm Right shin
Red and yellow Red Red
NA, but most likely
Actinic keratoses and melanoma in situ of the scalp
Fraga and Prossick (2010)70
62, F
NA
NA
KA
Ankle
Red
NA
NA
Fraga and Prossick (2010)70
66, M
NA
NA
KA
Leg
Red
NA
NA
Fraga and Prossick (2010)70
36, F
NA
NA
KA
Ankle
Red
NA
NA
Fraga and Prossick (2010)70
53, M
NA
NA
KA and SCC
Arm
Red
NA
NA
Fraga and Prossick (2010)70
52, M
NA
NA
KA
Arm
Red
NA
NA
Fraga and Prossick (2010)70
50, M
NA
NA
KA
Arm
Red
NA
NA
Fraga and Prossick (2010)70
49, M
NA
NA
KA
Arm
Red
NA
NA
Vitiello et al (2010)48
39, M
3 months
1 month/2 months
Eruptive KA (x8)
Left leg
Red
NA, African American NA origin
Gon Ados et al (2009)69
60, F
4 month
3 months/1 month
KA
Right leg
Red
NA
NA
Sarma et al (2010)47
79, M
NA/NA
Poorly differentiated SCC
Left forearm
Black
Yes, homeless and sun-damaged dermis
No
1 months
>50 years
KA=keratoacanthoma. SCC=squamous-cell carcinoma. NA=data not available specifically. M=male. F=female. BCC=basal-cell carcinoma. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Diagnosis made by shaving, no full thickness excision (pseudoepitheliomatous process highly suggestive of KA).
Table 3: Main characteristics of reported cases of squamous-cell carcinomas and keratoacanthomas arising in tattoos
Several cases of SCCs have been reported on tattoos,43–48 and Fraga and Prossick70 reported the largest series of 11 keratoacanthomas in tattoos. Clinicians should be careful not to classify all keratoacanthomas as SCCs. Two different entities should be considered: genuine, trauma-induced keratoacanthomas that occur rapidly after a tattooing procedure, usually within a week to a year, and classic SCC that usually develops years after tattooing. Indeed, the delay after procedure is crucial data; most of the reported cases of keratoacanthoma on a tattoo occurred within a year after tattooing.70,71 In only one case, the diagnosis of keratoacanthoma could be doubtful since the lesion occurred 7–8 years after tattooing.65 Pseudoepitheliomatous hyperplasia is also a e164
benign lesion that develops rapidly after tattooing;55–62 however, it usually presents as a verrucous lesion that can be difficult to distinguish from keratoacanthoma and SCC, clinically and pathologically.71 A full excision of these lesions (keratoacanthoma or pseudoepitheliomatous hyperplasia) is necessary to make the correct diagnosis. The characteristics of patients with a reported tumour arising in a tattoos are summarised in tables 1–5. Additional issues regarding the risk of skin cancers on tattoos include the tattooing procedure itself, and the tattoo masking the underlying skin. Indeed, tattooing on a naevus can traumatise it and trigger sudden clinical and pathological modifications prompting excision and microscopic analysis to rule out malignancy. Multiple www.thelancet.com/oncology Vol 13 April 2012
Review
tattoos can make clinical surveillance of naevi difficult. The ugly duckling sign (a mole that appears different from all others on the body) might become less visible when moles are surrounded by tattoos, and this situation is further complicated when a patient has a past history of melanoma or atypical mole syndrome. Dermoscopy is difficult since tattoo pigments appear as multiple blackblue globules, which can prevent accurate diagnosis or surveillance of lesions. All of these issues increase the risk of delayed diagnosis of a cutaneous malignancy.1
Schmidt and Christensen (1978)57§
Potential links between tattoos and skin cancers If there is a true link between tattooing and skin cancer, carcinogenesis is most likely to be the result of a multifactorial process. The trauma induced by the procedure (puncturing the skin) has typically been identified as one of the main causes. Tattoo pigments do not remain inert in the dermis; an inflammatory reaction occurs over the lifetime of the bearer, in an attempt to degrade all foreign material.74 Trauma, scars, or cutaneous chronic inflammation can underlie the development of BCCs, SCCs, and keratoacanthomas.35 Trauma specifically seems to be a crucial trigger for keratoacanthomas, since most cases occur within the first year after tattooing (table 4), whereas melanoma has not been proven to be directly associated with trauma.73 Also, the scarring process made by a tattoo is very different from other carcinogenic scars, such as a burn scar or chronic inflammation (eg, chronic hidradenitis). It is impossible to draw a direct conclusion from the carcinogenic process leading from a burn scar to SCC, and the potential carcinogenic effect of a tattoo.35 Chronic ultraviolet sun exposure has been suggested as another potential factor for skin carcinogenesis in tattoos. Tattoos might be located on ultraviolet-exposed or intermittently exposed body areas.36,39,40,42,44,45 Wiener and Scher38 suggested that pigments might alter ultraviolet absorption in the skin. Moreover, it could be postulated that a tattooed individual would show off their tattoos, especially during summer time, exposing the tattooed skin to sun and ultraviolet radiation with the potential long-term risks. However, sun habits and underlying skin condition are often not mentioned in reports, or are not accurately reported by patients. Among the reported cases of BCCs in tattoos, Wiener and Scher38 did not find any other lesion related to chronic sun exposure in a patient who was a merchant seaman; another patient applied regularly sunblock,39 and the youngest patient reported so far had no specific risk factor for non-melanoma skin cancer and a tattoo located on a non-sun-exposed location.40 Individuals with tattoos are usually aware that chronic sun exposure will alter the tattoo and cause colours to fade. Heavily tattooed individuals are reported to avoid sun exposure to protect their tattoos.74 Unfortunately, so far there has been no specific study on the sun habits of tattooed individuals. www.thelancet.com/oncology Vol 13 April 2012
Goldstein (1967)55†
Duration of Delay of onset after Tattoo Age tattooing/prior to location (years) at tattoo at diagnosis presentation diagnosis* and sex
Colour Chronic Other skin within sun affected exposure cancer area
NA
NA
NA
NA
Red
NA
NA
9 years
9 years/NA
Arm
Red
NA
NA
NA
NA
2–4 weeks/NA
NA
Red
NA
NA
Balfour et al (2003)58
27, F
1 year
2 months/10 months NA
Purple
NA
NA
Cui et al (2007)59
59, F
2 years
1 week/2 years
Leg
Red
NA
NA
Kluger et al (2008)60
30, F
7 months
1 month/6 months
Ankle
Red
NA
NA
Kluger et al (2008)60
32, F
NA
4 days/2 months
Upper back
Red
NA
NA
Kluger et al (2008)60
54, F
2·25 years
3 months/2 years
Ankle
Black
NA
NA
Then et al (2009)61
24, M
NA
1 year/NA
Leg
Red
NA
No
Biwas (2011)62
38, F
7 months
3 weeks/6 months
Ankle
Red
NA
NA
Biro and Klein 26, M (1967)56‡
NA=not available. M=male. F=female. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Author reported two cases from 15 individuals presenting with pseudoepitheliomatous reaction in a red tattoo. ‡Reaction occurred 6 weeks after the tattoo was lacerated by an accident at work. §Report of seven Danish prisoners who developed ulcerated and itchy skin tumours on red areas of tattoos, with pseudoepitheliomatous hyperplasia features overlying a granulomatous reaction.
Table 4: Main characteristics of reported cases of pseudoepitheliomatous hyperplasia arising in tattoos
Does the colour of the tattoo matter? Our review of case reports showed that most of the reported cases of melanoma and BCCs occurred on black, dark blue, or dark coloured tattoos (tables 1 and 2), whereas SCCs, keratoacanthomas, and benign pseudoepitheliomatous hyperplasia developed mainly within red tattoos (tables 3 and 4). Red and black have always been the most commonly used colours for tattoos.3 The high prevalence of such reactions in these colours might simply reflect the frequency of their use. Additionally, skin surveillance for malignancies is more difficult when dark tattoos are present. It is possible that suspicious lesions are detected and excised more rapidly when they are noted on a light coloured tattoo than on a darker tattoo. But such a statement does not explain the detection of melanoma and BCC on dark colours versus SCC and keratoacanthomas on red-coloured tattoos. It is possible that the PAHs detected in black inks have a role in the carcinogenesis of lesions. However, red is known to be the main colour responsible for hypersensitivity tattoo reaction,1 and according to our review, was the most common colour for occurrence of SCCs, keratoacanthomas, and pseudoepitheliomatous hyperplasia. Different compositions of inks might therefore have crucial roles for different pathogenic processes. Unfortunately, most studies have focused on black and e165
Review
Age (years) at diagnosis and sex
Duration of Delay of onset after tattooing/prior to tattoo at presentation diagnosis*
Sangeza et al (1992)52†
54, M
>20 years
Baker et al (2005)49
35, M
8 years
1 year/7 years
Reddy et al (2011)50‡
52, M
7 years†
West et al (2009)51
41, M
10 years
Diagnosis
Longstanding history Cutaneous B-cell of pseudolymphoma lymphoma reaction to tattoos
Tattoo location
Chronic Other skin Colour cancer sun within affected exposure area
Arm
Red
NA
Lymphoma also developed in tattoo-free area
Dermatofibrosarcoma Left forearm Black protuberans
NA
No
2 years/5 years
Dermatofibrosarcoma Right upper protuberans back
Red
NA
NA
9 years/1 year
Leiomyosarcoma
Black
NA
No
Right arm
M=male. NA=data not available specifically. *Delay of onset after tattooing=time span between tattoo application and occurence of skin tumour in the tattoo. Delay prior to diagnosis=time span between first noted occurence of skin lesion in the tattoo by the patient and the medical consultation that confirmed diagnosis. †Two previous cases of lymphoma on a tattoo have been reclassified as pseudolymphoma (a reactive benign polyclonal lymphocyte infiltration of the skin)53,54 ‡Patient had a previous tattoo 15 years before presentation on the same location, which was partially removed by laser 5 years later. A new tattoo was designed over the same site 7 years before presentation (ie, 2 years after laser removal of the original tattoo).
Table 5: Main characteristics of reported cases of miscellaneous tumours arising in tattoos
red inks. To our knowledge, except for the presence of metallic salts,6,7 the presence of potential carcinogenic components in other colours has not been investigated. Finally, there might be other undetermined factors, such as a genetic predisposition and viruses in the skin, with potential role in carcinogenesis associated with tattoos. Overall, the clinical relevance of malignancies in tattoos is unclear. In-vivo data are needed, such as analysis of ink composition at the site of cancer development compared with the composition on an unaffected area of the tattoo that contains the same colour. Compared with older case reports, it seems that in more recent reports patients who develop skin cancers on tattoos are younger and have a shorter delay since tattoo application.35 This observation could have two explanations. Skin cancer in younger people is unusual, but not exceptional nowadays. The younger age of patients with reported malignancies on tattoos might simply reflect this tendency. As tattooing becomes more popular, the likelihood of having a coincidental malignant lesion on a tattoo increases, and the risk is higher the more tattoos the individual has.35 Alternately, it could be speculated that the composition of ink is changing. Despite the presence of metallic salts such as mercury, getting a tattoo was possibly safer in the past than it is now (from a toxicological point of view), and new compounds with potential carcinogenic properties in recent inks might be responsible for the development of skin cancers.
Tattoo: a potential risk factor for inner organ cancers? Tattoo pigments reside mainly in the dermis, either within fibroblasts or between collagen bundles.74 Macrophages, in an attempt to eliminate the pigments, assimilate them and migrate to the lymphatic vessels then to the regional lymph nodes. Cases of macroscopic, often asymptomatic, lymph nodes filled with black pigments have been reported.75 As a generalisation, any individual with a tattoo might have a e166
regional black lymph node. Black lymph nodes can also become palpable, raising the concern for potential infection or malignancy and leading to lymph node excision75 or dissection by a surgeon who suspects metastatic melanoma. A diagnosis of metastastic melanoma might be difficult for a pathologist to assess because of the presence of exogenous pigments.76 Overall, the potential pathological relevance of a lymph node containing tattoo pigment has never been assessed. Engel and colleagues,17 using an SKH-I mouse model, noted that pigment concentration of PR 22 that was intradermally injected on mouse skin decreased by 30% within 6 weeks after tattooing. Lehner and colleagues20 extracted various red pigments (PR 9, 12, and 112) from tattooed skin of deceased patients (duration of tattoo unknown) and estimated a concentration of 0·002–0·110 mg/cm². They estimated that the remaining pigment concentration in these red tattoos was 1–13% of the original concentration.20 The precise kinetics of the decrease over time of tattoo pigments remains to be assessed on standardised human skin samples. This decrease is due to elimination of some of the pigment during the healing phase, decomposition by repeated solar radiation, and migration into the lymph nodes. So far, there is no data regarding the potential localisation of tattoo byproducts elsewhere in the body. Nevertheless, some researchers stress that the long contact time between skin and tattoos allows continued exposure to toxic or allergenic metals with consequent accumulation in the body.7 The potential diffusion of nanoparticles also remains speculative.19 Tattoo pigments might be discharged from the body without any further health-related complications.
Conclusions With roughly 50 case reports of cutaneous malignancies on tattoos over the past 40 years, and millions of tattooed individuals worldwide, the number of cases remains at the level of background noise. So far, this association has to be considered as coincidental. www.thelancet.com/oncology Vol 13 April 2012
Review
Search strategy and selection criteria References for this Review were found through a search of PubMed by use of the terms “tattoo”, “tattoos”, or “tattooing”. Since 2006, NK has collected articles on health-related issues associated with tattooing, by performing a prospective weekly search of PubMed using the above keywords. An additional search was done using the terms “tattoo or tattoos or tattooing” AND “cancer”, “skin cancer”, “melanoma”, “squamous-cell carcinoma”, “basal-cell carcinoma”, “keratoacanthoma”, “pseudo-epitheliomatous hyperplasia”, “lymphoma”, or “sarcoma”. Reference lists of relevant articles were reviewed. One conference abstract was included because it related directly to the topic. No date restrictions were used. Only articles published in English and French were included. Selected references on associated topics were also included.
6 7 8 9
10
11
12
13
14
Unfortunately, so far only a few groups are investigating the association between tattoos and skin tumours, and large-scale studies of clinical and epidemiological factors are lacking. Additional, in-vivo investigations of cutaneous concentrations of pigments and byproducts are needed. Large cohort follow-up of tattooed individuals would help to assess whether or not tattooing is an independent risk factor for melanoma, non-melanoma skin cancer, and visceral cancers. The addition of tattoo data in national skin cancer registries would help define the true prevalence of malignancies in tattoos. There are no data available regarding the sun habits of tattooed individuals, who are thought to avoid sun to protect their tattoos. Permanent tattooing with its health-related issues and potential carcinogenic effect is a topic that should not be neglected by health services, and could be an important upcoming health issue. Regulation of tattoo ink manufacturing is urgently needed. Contributors NK had the original idea for the article. Both authors contributed to conception and design of the report, and the acquisition, analysis, and interpretation of data. Both did the literature search, wrote and edited the report, and approved it for publication.
15
16
17
18
19 20
21 22 23 24 25 26
Conflicts of interest We declare that we have no conflicts of interest. References 1 Kluger N. Cutaneous complications related to permanent decorative tatooing. Expert Rev Clin Immunol 2010; 6: 363–71. 2 Laumann AE, Derick AJ. Tattoos and body piercings in the United States: a national data set. J Am Acad Dermatol 2006; 55: 413–21. 3 Klügl I, Hiller KA, Landthaler M, Bäumler W. Incidence of health problems associated with tattooed skin: a nation-wide survey in German-speaking countries. Dermatology 2010; 221: 43–50. 4 Bäumler W, Vasold R, Lundsgaard J, Talberg HJ. Chemical used in tattooing and permanent make up products. In: Papameletiou D, Schwela D, Zenie A, eds. Workshop on Technical/Scientific and Regulatory Issues on the Safety of Tattoos, Body Piercing and of Related Practices. Ispra, VA: European Commission, 2003: 21–36. 5 Lehner K, Santarelli F, Vasold R, König B, Landthaler M, Bäumler W. Black tattoo inks are a source of problematic substances such as dibutyl phthalate. Contact Dermatitis 2011; 65: 231–38.
www.thelancet.com/oncology Vol 13 April 2012
27 28 29
30 31 32
33
Beute TC, Miller CH, Timko AL, Ross EV. In vitro spectral analysis of tattoo pigments. Dermatol Surg 2008; 34: 508–15. Forte G, Petrucci F, Cristaudo A, Bocca B. Market survey on toxic metals contained in tattoo inks. Sci Total Environ 2009; 407: 5997–6002. Vasold R, Engel E, König B, Landthaler M, Bäumler W. Health risks of tattoo colors. Anal Bioanal Chem 2008; 391: 9–13. Cui Y, Spann AP, Couch LH, et al. Photodecomposition of Pigment Yellow 74, a pigment used in tattoo inks. Photochem Photobiol 2004; 80: 175–84. Cui Y, Churchwell MI, Couch LH, Doerge DR, Howard PC. Metabolism of pigment yellow 74 by rat and human microsomal proteins. Drug Metab Dispos 2005; 33: 1459–65. Vasold R, Naarmann N, Ulrich H, et al. Tattoo pigments are cleaved by laser light-the chemical analysis in vitro provide evidence for hazardous compounds. Photochem Photobiol 2004; 80: 185–90. Engel E, Santarelli F, Vasold R, et al. Establishment of an extraction method for the recovery of tattoo pigments from human skin using HPLC diode array detector technology. Anal Chem 2006; 78: 6440–47. Engel E, Spannberger A, Vasold R, König B, Landthaler M, Bäumler W. Photochemical cleavage of a tattoo pigment by UVB radiation or natural sunlight. J Dtsch Dermatol Ges 2007; 5: 583–89. Wang L, Yan J, Hardy W, Mosley C, Wang S, Yu H. Light-induced mutagenicity in Salmonella TA102 and genotoxicity/cytotoxicity in human T-cells by 3,3´-dichlorobenzidine: a chemical used in the manufacture of dyes and pigments and in tattoo inks. Toxicology 2005; 207: 411–18. Engel E, Santarelli F, Vasold R, et al. Modern tattoos cause high concentrations of hazardous pigments in skin. Contact Dermatitis 2008; 58: 228–33. Gopee NV, Cui Y, Olson G, et al. Response of mouse skin to tattooing: use of SKH-1 mice as a surrogate modelfor human tattooing. Toxicol Appl Pharmacol 2005; 209: 145–58. Engel E, Vasold R, Santarelli F, et al. Tattooing of skin results in transportation and light-induced decomposition of tattoo pigments—a first quantification in vivo using a mouse model. Exp Dermatol 2010; 19: 54–60. Regensburger J, Lehner K, Maisch T, et al. Tattoo inks contain polycyclic aromatic hydrocarbons that additionally generate deleterious singlet oxygen. Exp Dermatol 2010; 19: 275–81. Høgsberg T, Loeschner K, Löf D, Serup J. Tattoo inks of general usage contain nanoparticles. Br J Dermatol 2011; 165: 1210–18. Lehner K, Santarelli F, Penning R, et al. The decrease of pigment concentration in red tattooed skin years after tattooing. J Eur Acad Dermatol Venereol 2011; 25: 1340–45. Sharlit H. Melanoma caused by indelible pencil. Arch Dermatol 1938; 37: 301–06. Allen AC. The skin. St Louis: CV Mosby Co, 1954. Kirsch N. Malignant melanoma developing in a tattoo. Arch Dermatol 1969; 99: 596–98. Kirsch N. Malignant melanoma developing in a tattoo. Int J Dermatol 1972; 11: 16–20. Wolfort FC, Hoopes JE, Filtze HS, Cochran TC. Superficial melanoma in a tattoo. Br J Plast Surg 1974; 27: 303–04. Bartal AH, Cohen Y, Robinson E. Malignant melanoma arising at tattoo sites used for radiotherapy field marking. Br J Radiol 1980; 3: 913–14. Lee YT, Craig JR. Melanoma in a tattoo of the breast. J Surg Oncol 1984; 25: 100–01. Kircik L, Armus S, van den Broek H. Malignant melanoma in a tattoo. Int J Dermatol 1993; 32: 297–98. Soroush V, Gurevitch AW, Peng SK. Malignant melanoma in a tattoo: case report and review of the literature. Cutis 1997; 59: 111–12. Khan IU, Moiemen NS, Firth J, Frame JD. Malignant melanoma disguised by a tattoo. Br J Plast Surg 1999; 52: 598. Stinco G, De Francesco V, Frattasio A, Quinkenstein E, Patrone P. Malignant melanoma in a tattoo. Dermatology 2003; 206: 345–46. Paradisi A, Capizzi R, De Simone C, Fossati B, Proietti I, Amerio PL. Malignant melanoma in a tattoo: case report and review of the literature. Melanoma Res 2006; 16: 375–76. Shariff Z, Tehrani H, Jagadeesan J, Hardwicke J. Artwork: to be studied. Dermatol Online J 2006; 12: 21.
e167
Review
34
35
36 37 38 39 40 41 42 43 44
45 46
47
48
49 50
51
52
53 54
55
56
e168
Singh RS, Hafeez Diwan A, Prieto VG. Potential diagnostic pitfalls in melanoma arising in a cutaneous tattoo. Histopathology 2007; 51: 283–85. Kluger N, Phan A, Debarbieux S, Balme L, Thomas L. Skin cancers arising in tattoos: coincidental or not? Dermatology 2008; 217: 219–21. Bashir AH. Basal cell carcinoma in tattoos: report of two cases. Br J Plast Surg 1976; 29: 288–90. Earley MJ. Basal cell carcinoma arising in tattoos: a clinical report of two cases. Br J Plast Surg 1983; 36: 258–59. Wiener DA, Scher RK. Basal cell carcinoma arising in a tattoo. Cutis 1987; 39: 125–26. Doumat F, Kaise W, Barbaud A, Schmutz JL. Basal cell carcinoma in a tattoo. Dermatology 2004; 208: 181–82. Birnie AJ, Kulkarni K, Varma S. Basal cell carcinoma arising in a tattoo. Clin Exp Dermatol 2006; 31: 820–21. Lee JS, Park J, Kim SM, Yun SK, Kim HU. Basal cell carcinoma arising in a tattooed eyebrow. Ann Dermatol 2009; 21: 281–84. Omidian M, Emad-Mostofi N. Basal cell carcinoma arising from traditional tattoo. Arch Iranian Med 2009; 12: 198. MacQuarrie DG. An unusual breast cancer: squamous cell carcinoma arising in a tattoo. Minn Med 1966; 49: 799–801. Pitarch G, Martínez-Menchón T, Martínez-Aparicio A, Sánchez-Carazo JL, Muñoz D, Fortea JM. Squamous cell carcinoma over tattoos. J Am Acad Dermatol 2007; 56: 1072–73. Tan-Billet J, Khan U, Roberson D, Rao B. Squamous cell carcinoma developing in a tattoo. J Am Acad Dermatol 2007; 56 (suppl 2): 150. Ortiz A, Yamauchi PS. Rapidly growing squamous cell carcinoma from permanent makeup tattoo. J Am Acad Dermatol 2009; 60: 1073–74. Sarma DP, Dentlinger RB, Forystek AM, Stevens T, Huerter C. Poorly differentiated squamous cell carcinoma arising in tattooed skin. Case Report Med 2010; 2010: 431813. Vitiello M, Echeverria B, Romanelli P, Abuchar A, Kerdel F. Multiple eruptive keratoacanthomas arising in a tattoo. J Clin Aesthet Dermatol 2010; 3: 54–55. Baker PA, O’Dowd GJ, Khan IU. Dermatofibrosarcoma protuberans arising in a decorative tattoo. Sarcoma 2005; 9: 37–41. Reddy KK, Hanke CW, Tierney EP. Malignancy arising within cutaneous tattoos: case of dermatofibrosarcoma protuberans and review of literature. J Drugs Dermatol 2011; 10: 837–42. West CC, Morritt AN, Pedelty L, Lam DG. Cutaneous leiomyosarcoma arising in a tattoo—‘a tumour with no humour’. J Plast Reconstr Aesthet Surg 2009; 62: 79–80. Sangueza OP, Yadav S, White CR Jr, Braziel RM. Evolution of B-cell lymphoma from pseudolymphoma. A multidisciplinary approach using histology, immunohistochemistry, and Southern blot analysis. Am J Dermatopathol 1992; 14: 408–13. Auger C. Cancer sur tatouage. Laval Med 1943; 8: 300. Armiger WG, Caldwell EH. Primary lesion of a non-Hodgkin’s lymphoma occurring in a skin tatoo: case report. Plast Reconstr Surg 1978; 62: 125–27. Goldstein N. Mercury-cadmium sensitivity in tattoos. A photoallergic reaction in red pigment. Ann Intern Med 1967; 67: 984–89. Biro L, Klein WP. Unusual complications of mercurial (cinnabar) tattoo. Generalized eczematous eruption following laceration of a tattoo. Arch Dermatol 1967; 96: 165–67.
57 58
59
60
61 62 63 64 65
66
67 68
69 70
71
72
73
74
75 76
Schmidt H, Christensen HE. Red poster paint tattoo granuloma. Arch Dermatol 1978; 114: 965–66. Balfour E, Olhoffer I, Leffell D, Handerson T. Massive pseudoepitheliomatous hyperplasia: an unusual reaction to a tattoo. Am J Dermatopathol 2003; 25: 338–40. Cui W, McGregor DH, Stark SP, Ulusarac O, Mathur SC. Pseudoepitheliomatous hyperplasia—an unusual reaction following tattoo: report of a case and review of the literature. Int J Dermatol 2007; 46: 743–45. Kluger N, Durand L, Minier-Thoumin C, et al. Pseudoepitheliomatous epidermal hyperplasia in tattoos: report of three cases. Am J Clin Dermatol 2008; 9: 337–40. Then M, Mark Boustred A, Clarke LE. Keratoacanthomatous hyperplasia in response to a tattoo. Dermatol Surg 2009; 35: 685–86. Biswas A. Pseudoepitheliomatous tattoo reaction. Diagnostic Histopathology 2011; 17: 272–75. Cipollaro VA. Keratoacanthoma developing in a tattoo. Cutis 1973; 11: 809–10. Goldstein N. IV. Complications from tattoos. J Dermatol Surg Oncol 1979; 5: 869–78. Kleinerman R, Greenspan A, Hale EK. Mohs micrographic surgery for an unusual case of keratoacanthoma arising from a longstanding tattoo. J Drugs Dermatol 2007; 6: 931–32. Kluger N, Minier-Thoumin C, Plantier F. Keratoacanthoma occurring within the red dye of a tattoo. J Cutan Pathol 2008; 35: 504–07. Chorny JA, Stephens FV, Cohen JL. Eruptive keratoacanthomas in a new tattoo. Arch Dermatol 2007; 143: 1457–58. Goldenberg G, Patel S, Patel MJ, Williford P, Sanguenza O. Eruptive squamous cell carcinomas, keratoacanthoma type, arising in a multicolor tattoo. J Cutan Pathol 2008; 35: 62–64. Gon Ados S, Minelli L, Meissner MC. Keratoacanthoma in a tattoo. Dermatol Online J 2009; 15: 9. Fraga GR, Prossick TA. Tattoo-associated keratoacanthomas: a series of 8 patients with 11 keratoacanthomas. J Cutan Pathol 2010; 37: 85–90. Kluger N. Issues with keratoacanthoma, pseudoepitheliomatous hyperplasia and squamous cell carcinoma within tattoos: a clinical point of view. J Cutan Pathol 2010; 37: 812–13. Varga E, Korom I, Varga J, Kohán J, Kemény L, Oláh J. Melanoma and melanocytic nevi in decorative tattoos: three case reports. J Cutan Pathol 2011; 38: 994–98. Kaskel P, Kind P, Sander S, Peter RU, Krähn G. Trauma and melanoma formation: a true association? Br J Dermatol 2000; 143: 749–53. Kluger N, Plantier F, Moguelet P, Fraitag S. Tattoos: natural history and histopathology of cutaneous reactions. Ann Dermatol Venereol 2011; 138: 146–54. Kluger N, Cohen-Valensi R, Nezri M. Black lymph nodes—and a colourful skin. Lancet 2008; 371: 1214. Jaigirdar AA, Yeh MW, Sharifi E, Browne LW, Leong SP. Coexisting tattoo pigment and metastatic melanoma in the same sentinel lymph node. J Cutan Med Surg 2009; 13: 321–25.
www.thelancet.com/oncology Vol 13 April 2012