Predisposing factors of actinic keratosis in a North-West German population

ARTICLE

Auteur(s) : Peter Hensen¹, Marcel L Müller², Ramin Haschemi¹, Hartmut Ständer³, Thomas A Luger¹, Cord Sunderkötter¹, Meinhard Schiller¹

¹Department of Dermatology, University of Münster, Von-Esmarch-Str. 58, D-48149 Münster, Germany
²Department of Dermatology, University of Freiburg, D-79104 Freiburg, Germany
³Dermatology Bad Bentheim, Paulinenkrankenhaus, Bad Bentheim, Germany

accepté le 12 Mars 2009

Skin cancer has become an important public health problem among Caucasians with a steadily increasing incidence in recent decades [1-3]. The majority of these are non-melanoma skin cancers, i.e. basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). While the incidence rates of invasive non-melanoma skin cancers are already alarming, i.e. 20-30 SCC and100 BCC new cases/100,000/year respectively, in central Europe [4], the number of carcinomas in situ of SCC, also referred to as actinic keratosis (AK), is excessively high (approximately 250 new cases/100,000/year) [5]. The prevalence of this entity in Europe is 15% in men and 6% in woman [6]. Over the age of 70, as many as 34% of males and 18% of females were found to have AK in Europe. The USA shows a prevalence between 11-26% [7], while the highest prevalence has been reported for Australia (Queensland) where 55% of men between 30-70 years showed AK [8, 9]. The forecast for future non-melanoma skin cancer incidence rates is dramatic [3, 10].

AK is an initial manifestation of a continuum of clinical and histological abnormalities that progresses from AK to invasive SCC. Approximately 10% of AK reveal such a progression to invasive SCC) [11, 12]. The percentage is dramatically higher in patients under immunosuppression [13], usually patients who have received organ transplantation. The term keratinocytic intraepidermal neoplasia has been suggested to replace the term actinic keratosis to make it clear that AK is a genuine neoplasm (carcinoma in situ) which has a potential to break beyond its epidermal confinement into the dermis, [14, 15].

AK are most commonly found on chronically sun-exposed regions of the skin. They present as skin colored to reddish brown or yellowish black thin or raised papules or plaques with discrete keratosis (appearing like dry adherent scale), sometimes also with marked or even horn-like keratosis [16]. The diagnosis of AK can usually be made clinically by experienced dermatologists. Histological confirmations revealed a positive predictive value range from 74% up to 94% [17-19]. AK in children are limited to rare genetic disorders such as albinism and xeroderma pigmentosum [20]. The prevalence of AK increases highly with advancing age [21-23]. Furthermore, the prevalence of AK was found to be significantly higher in men than in women [24], which corresponds with findings on the prevalence or incidence of other non-melanoma skin cancers in Europe [25, 26]. High levels of UV exposure have been closely associated with the appearance of AK in at least some studies, and prevalence has been strongly related to individual skin sensitivity and objective signs of sun exposure such as solar elastosis and solar lentigines in some studies [6, 8, 27]. Available data on the epidemiology of AK and possible predisposing factors refer commonly to populations in Australia or the United States, as the majority of the studies have been conducted in these countries. By contrast, available epidemiologic data on AK for Central Europe are sometimes controversial and limited [24, 28]. This is remarkable because these areas are densely populated so that data on risk factors will have a high socioeconomic impact. The results might vary from the studies in Australia or the US because sun hours and thus sun exposure is lower in this geographic area. The aim of this study was therefore to assess, as well as to compare, clinical, demographic and behavioral factors of patients with and without AK from a prototypical fair skinned Central European population, and to detect possible factors that are predisposing for the appearance of AK in this area.

Methods

Study design

Cases and controls were recruited as follows: At the participating outpatient centers, both cases and controls were continuously recruited over the investigation period. Cases were selected on the basis of clinical diagnosis of AK, because clinical recognition of AK by experienced specialists has been demonstrated to have a high positive predictive value for diagnostic accuracy [17-19]. Data on inter-observer variability were not collected. To recruit control subjects we randomly selected patients admitted to the outpatient center with conditions not suspected of being related to neoplastic diseases. Controls were included in this study only if full clinical examination by trained dermatologists revealed no sign of an AK or of any other suspicious lesions related to skin cancer. Questionnaires were handed out to the control subjects selected with a short declaration of the purpose of the study. In the dermatological department, cases were selected if a diagnosis of AK was made clinically in affected inpatients or outpatients and if the diagnosis could be histologically confirmed. Names and addresses were retrieved from the patient data management system and questionnaires were sent out. Control subjects were recruited in the same manner in the outpatient centers. The questionnaire developed for this study contained 23 items which had been suggested according to results from previous research and they were validated in internal audits among oncology specialists. To increase feasibility and response-rates, the questionnaire size was limited to two pages and the number of items was restricted to a reasonable quantity. Each item was carefully worded in a clear and precise manner. Accurate item understanding was tested on a pre-test with elderly people. Items concerned demographic data (e.g. age and sex), personal characteristics (e.g. skin phototype and color of hair), relevant personal habits (e.g. use of sunscreen or smoking), history of dermatological disorders and chronic diseases (e.g. psoriasis and warts), personal and family history of skin malignancies, transplantations undergone, history of sun exposure for recreational and occupational reasons, history of sunburn in youth and adulthood, and, except for controls, the number and location of presenting AK lesions. Arsenic therapy, organ transplantation, skin warts, vitiligo, and psoriasis were included because they are discussed as possible risk factors for AK [20, 20, 29-32]. Data regarding socioeconomic status or a family history of skin cancer were not collected. The questionnaires were either sent by regular mail or were handed out directly by attending physicians. The whole study was performed anonymously. Therefore, reminding of non-respondents could not be undertaken. Furthermore, no incentives for completion were provided, but to avoid any additional expenses, respondents were provided with an accompanying stamped and pre-addressed envelope.

Data analysis

Results

Demographic data

Clinical data

Data from bivariate analysis

Data from multivariate analysis

^bAdjusted Odds Ratio has been calculated using raw age.

Risk pattern

Probability curves

Discussion

This study thus encompasses a comprehensive multivariate analysis probably of most, if not all the possible risk factors for AK in central Europe. Previous studies had already analyzed a variety of aspects of these risk factors, which were exploited by, but also complemented by this study. Some of these studies had looked at populations living under quite different climatic conditions. For example, sun hours and thus sun exposure differs significantly in Australia or the US compared to central Europe and thus might have a different capacity to initiate AK growth.

The role of chronic UV radiation in non-melanoma skin cancer, i.e. BCC and SCC, is well defined [34-37]. However, whether these findings are applicable to the concept of AK has been less thoroughly examined. A first clue comes from the fact that the distribution of AK closely correlates with the body sites of most intense chronic ultraviolet light exposure. Correspondingly, we found that AK are predominately located on the face and head region, i.e. 86.1% of the cases reported having AK on this site (table 1). Nevertheless, a possible selection bias could be involved, as the high prevalence rate on the face and head- body areas which benefit from a daily self-inspection – might be overestimated. So far, results concerning the contribution of occupational vs. recreational sun exposure are not consistent. A recent study investigating associated factors of AK in a representative Italian sample undoubtedly verified that the prevalence of AK increased with the number of hours spent in the sun during the week and on holiday [24]. Similarly, a case control study from the Netherlands reported that the cumulative sun exposure correlated with an increased risk of actinic keratosis [28]. These two European studies, however, did not differentiate between occupational and recreational sun exposure [24, 28].

In our study, sun exposure for occupational reasons was especially found to be part of a risk profile for a north-western German population. However, as excessive lifetime sun exposure rates are generally only experienced by outdoor workers [38, 39], it is not surprising that several studies from Australia and the US demonstrated high incidence rates for AK in people with such professions [8, 39, 40]. Working outdoors was also found in Japan to be a risk factor for precancerous lesions, including AK [41]. However, Memon et al. could only detect a slightly higher, non-significant prevalence of AK in subjects who had worked outdoors at four centers in the northwest of England (Mersey region) [6]. Thus, there is a discrepancy in the data of other studies on the contribution of occupational sun exposure; since Memon et al. did not perform multivariate analysis [6] confounding factors such as a high portion of subjects with a Celtic phenotype within the study population could have influenced the outcome of the analysis. As men are more prone to have outdoor occupations, it is not surprisingly that men are more likely than women to develop AK in our study, in accordance with the literature [6, 8, 28, 41, 42]. This fact may serve as further evidence that work-related sun exposure is a crucial AK risk factor.

Furthermore, we identified by multivariate analysis that insufficient use or denial of use of sunscreens is a weak but independent variable predicting the risk of AK. At first sight, this positive association appeared predictable. However, two previous studies did not find such a correlation [8, 24]. Frost et al. reported that in adults older than 20 years sun protection practices were even positively associated with the occurrence of AK [8, 9]. This was explained by the presence of residual confounding factors such as skin phenotype and prolonged sun exposure, most likely due to a false feeling of security [8, 9]. Two other prospective studies – one from Australia [18] and one from Japan [42] – were in line with our findings as they revealed that regular use of sunscreen with a sun protection factor of 17 would indeed prevent the development of AK [18], and sunscreen use with age (> 60 years) decreased new cases of AK among the Japanese [42]. Our study confirmed that the recall of painful sunburns before the age of 20 y is associated with an increased risk of AK [8, 9, 28], as has been demonstrated for SCC, BCC and melanoma [43, 44]. These early sunburns may be considered to be initiating events in the multistep process of developing skin cancer [45]. Acute and chronic sun ultraviolet radiation of the skin lead to mutations in the p53 gene and subsequently to clonal expansion of mutated keratinocytes, which is manifested clinically as the development of actinic keratosis [46]. Functional loss of th p53 tumour suppressor gene together with mutations of RAS and certain chromosomal aberrations could finally lead to squamous cell carcinoma. Therefore it is not surprising that several studies showed significant associations between a history of previous non-melanoma skin cancers and the presence of AK [24, 28] and SCC [37]. We demonstrated for the first time that a personal history of any skin malignancies is one of the strongest independent factors determining the development of actinic keratosis. Besides the individual lifetime sun exposure, a personal history of skin malignancies might reflect genetic factors that result in constitutional sensitivity to sunlight. Consequently, we demonstrated in accordance with literature [6, 8, 28, 40, 41] that pale phototype, freckles, colour of hair and eyes carried an increased risk ofdeveloping AK.

Case-control studies have some clear limitations. First, the findings are based on a limited sample set based on self-reported behavioural risk factors and personal characteristics. Second, the control subjects were selected from an identical clinical setting. Therefore it remains unclear whether the results of the study would have been similar if controls had been recruited in a setting outside the health care system. Third, transfer of results to other populations always requires critical assessment and the results of this study refer primarily to a Westphalian and Lower-Saxonian population.

The summarized risk pattern established here for the probability of developing an AK may serve skin cancer prevention programs as a basis for the improvement of clinical strategies and behavioural purposes. A special emphasis should made to increase the knowledge base of susceptible individuals, including people with blue eyes, fair hair and freckles, and immunocompromised patients [36]. Reduction or prevention of exposure, in particular avoidance of the sun, appears to be the paramount promising strategy to reduce the risk of AK. However, avoidance of excessive sun exposure may be especially difficult for outdoor workers. At least, they should be advised to use sunscreen daily [47, 48]. Because painful sunburns before the age of 20 are an additional risk factor for AK, reducing sunlight exposure in childhood may considerably decrease the incidence of AK and SCCs in later life. Furthermore, patients presenting the proposed risk pattern have to be thoroughly and regularly examined with regard to AK lesions. Despite the results summarized in this study and the increased public awareness of UV-induced health risks, recent studies have revealed that total body screening examinations are still infrequently performed, even among patients at increased risk because of excessive occupational sun exposure [49]. Although AK were found to be predominantly located on the face and head and, thus, appear to be readily identifiable, full body inspections by an experienced dermatologist are highly recommended. The threshold for a biopsy of suspected lesions, especially in patients with a history of skin cancer, should be low, to exclude early invasive SCC and BCC [19, 50]. We thus need to focus on primary and secondary strategies for prevention of skin cancer by the implementation of risk profiles into health practice. This way it is feasible to reduce incidence rates of skin cancer and to secure their early detection which in turn allows for less traumatizing therapeutic procedures to cure these early stages of skin cancer. Further research and deployment programs at the health system level are needed to implement prevention strategies much more into clinical practice and to increase the awareness of early skin cancer and safe sun practices among the Central European populations.

Acknowledgements

References

2 Miller DL, Weinstock MA. Nonmelanoma skin cancer in the United States: incidence. J Am Acad Dermatol 1994; 30: 774-8.

3 de Vries E, van de Poll-Franse LV, Louwman WJ, De Gruijl FR, Coebergh JW. Predictions of skin cancer incidence in the Netherlands up to 2015. Br J Dermatol 2005; 152: 481-8.

4 Garbe C. Epidemiologie des Hautkrebses. In: Garbe C, Dummer R, Kaufmann R, Tilgen W, eds. Dermatologische Onkologie. Berlin, Heidelberg, New York: Springer, 1997: 40-56.

5 Suzuki T, Ueda M, Naruse K, et al. Incidence of actinic keratosis of Japanese in Kasai City, Hyogo. J Dermatol Sci 1997; 16: 74-8.

6 Memon AA, Tomenson JA, Bothwell J, Friedmann PS. Prevalence of solar damage and actinic keratosis in a Merseyside population. Br J Dermatol 2000; 142: 1154-9.

7 Salasche SJ. Epidemiology of actinic keratoses and squamous cell carcinoma. J Am Acad Dermatol 2000; 42: 4-7.

8 Frost CA, Green AC, Williams GM. The prevalence and determinants of solar keratoses at a subtropical latitude (Queensland, Australia). Br J Dermatol 1998; 139: 1033-9.

9 Frost C, Williams G, Green A. High incidence and regression rates of solar keratoses in a queensland community. J Invest Dermatol 2000; 115: 273-7.

10 Hoey SE, Devereux CE, Murray L, et al. Skin cancer trends in Northern Ireland and consequences for provision of dermatology services. Br J Dermatol 2007; 156: 1301-7.

11 Glogau RG. The risk of progression to invasive disease. J Am Acad Dermatol 2000; 42: 23-4.

12 Quatresooz P, Pierard-Franchimont C, Paquet P, Hubert P, Delvenne P, Pierard GE. Crossroads between actinic keratosis and squamous cell carcinoma, and novel pharmacological issues. Eur J Dermatol 2008; 18: 6-10.

13 Stockfleth E, Ulrich C, Meyer T, Christophers E. Epithelial malignancies in organ transplant patients: clinical presentation and new methods of treatment. Recent Results Cancer Res 2002; 160: 251-8.

14 Heaphy Jr. MR, Ackerman AB. The nature of solar keratosis: a critical review in historical perspective. J Am Acad Dermatol 2000; 43: 138-50.

15 Cohn BA. From sunlight to actinic keratosis to squamous cell carcinoma. J Am Acad Dermatol 2000; 42: 143-4.

16 Moy RL. Clinical presentation of actinic keratoses and squamous cell carcinoma. J Am Acad Dermatol 2000; 42: 8-10.

17 Ponsford MW, Goodman G, Marks R. The prevalence and accuracy of diagnosis of non-melanotic skin cancer in Victoria. Australas J Dermatol 1983; 24: 79-82.

18 Thompson SC, Jolley D, Marks R. Reduction of solar keratoses by regular sunscreen use. N Engl J Med 1993; 329: 1147-51.

19 Venna SS, Lee D, Stadecker MJ, Rogers GS. Clinical recognition of actinic keratoses in a high-risk population: how good are we? Arch Dermatol 2005; 141: 507-9.

20 Schwartz RA. Premalignant keratinocytic neoplasms. J Am Acad Dermatol 1996; 35: 223-42.

21 Harvey I, Frankel S, Marks R, Shalom D, Nolan-Farrell M. Non-melanoma skin cancer and solar keratoses II analytical results of the South Wales Skin Cancer Study. Br J Cancer 1996; 74: 1308-12.

22 Marks R. Non-melanoma skin cancer and solar keratoses in Australia – a review. Eur J Epidemiol 1985; 1: 319-22.

23 Schwartz RA. The actinic keratosis. A perspective and update. Dermatol Surg 1997; 23: 1009-19.

24 Naldi L, Chatenoud L, Piccitto R, Colombo P, Placchesi EB, La Vecchia C. Prevalence of actinic keratoses and associated factors in a representative sample of the Italian adult population: Results from the Prevalence of Actinic Keratoses Italian Study, 2003-2004. Arch Dermatol 2006; 142: 722-6.

25 Roberts DL. Incidence of non-melanoma skin cancer in West Glamorgan, South Wales. Br J Dermatol 1990; 122: 399-403.

26 Hannuksela-Svahn A, Pukkala E, Karvonen J. Basal cell skin carcinoma and other nonmelanoma skin cancers in Finland from 1956 through 1995. Arch Dermatol 1999; 135: 781-6.

27 Green A, Whiteman D, Frost C, Battistutta D. Sun exposure, skin cancers and related skin conditions. J Epidemiol 1999; 9: S7-S13.

28 Kennedy C, Bajdik CD, Willemze R, De Gruijl FR, Bouwes Bavinck JN. The influence of painful sunburns and lifetime sun exposure on the risk of actinic keratoses, seborrheic warts, melanocytic nevi, atypical nevi, and skin cancer. J Invest Dermatol 2003; 120: 1087-93.

29 Ulrich C, Schmook T, Nindl I, Meyer T, Sterry W, Stockfleth E. Cutaneous precancers in organ transplant recipients: an old enemy in a new surrounding. Br J Dermatol 2003; 149 (Suppl 66): 40-2.

30 Forslund O, Ly H, Reid C, Higgins G. A broad spectrum of human papillomavirus types is present in the skin of Australian patients with nonmelanoma skin cancers and solar keratosis. Br J Dermatol 2003; 149: 64-73.

31 Seo SL, Kim IH. Squamous cell carcinoma in a patient with generalized vitiligo. J Am Acad Dermatol 2001; 45: S227-S2S9.

32 Paltiel O, Adler B, Herschko K, Tsukrov B, David M. Are patients with psoriasis susceptible to the classic risk factors for actinic keratoses? Arch Dermatol 2004; 140: 805-10.

33 Marshall T. What is a case-control study? Int J Epidemiol 2004; 33: 612-3.

34 Walther U, Kron M, Sander S, et al. Risk and protective factors for sporadic basal cell carcinoma: results of a two-centre case-control study in southern Germany. Clinical actinic elastosis may be a protective factor. Br J Dermatol 2004; 151: 170-8.

35 Ruiz LA, Kuznitzky R, Garay I, Ducasse C, Albertini R. Risk factors for basal cell carcinoma. Case-control study in Cordoba. Medicina (B Aires) 2005; 65: 495-500; (B Aires).

36 Ulrich C, Schmook T, Sachse MM, Sterry W, Stockfleth E. Comparative epidemiology and pathogenic factors for nonmelanoma skin cancer in organ transplant patients. Dermatol Surg 2004; 30: 622-7.

37 Han J, Colditz GA, Hunter DJ. Risk factors for skin cancers: a nested casecontrol study within the Nurses’ Health Study. Int J Epidemiol 2006; 35: 1514-21.

38 Vishvakarman D, Wong JC, Boreham BW. Annual occupational exposure to ultraviolet radiation in central Queensland. Health Phys 2001; 81: 536-44.

39 Schwartz RA, Bridges TM, Butani AK, Ehrlich A. Actinic keratosis: an occupational and environmental disorder. J Eur Acad Dermatol Venereol 2008; 22: 606-15.

40 Vitasa BC, Taylor HR, Strickland PT, et al. Association of nonmelanoma skin cancer and actinic keratosis with cumulative solar ultraviolet exposure in Maryland watermen. Cancer 1990; 65: 2811-7.

41 Naruse K, Ueda M, Nagano T, et al. Prevalence of actinic keratosis in Japan. J Dermatol Sci 1997; 15: 183-7.

42 Araki K, Nagano T, Ueda M, et al. Incidence of skin cancers and precancerous lesions in Japanese – risk factors and prevention. J Epidemiol 1999; 9: S14-S21.

43 Leiter U, Garbe C. Epidemiology of melanoma and nonmelanoma skin cancer-the role of sunlight. Adv Exp Med Biol 2008; 624: 89-103.

44 Armstrong BK, Kricker A. The epidemiology of UV induced skin cancer. J Photochem Photobiol B 2001; 63: 8-18.

45 Kubo Y, Murao K, Matsumoto K, Arase S. Molecular carcinogenesis of squamous cell carcinomas of the skin. J Med Invest 2002; 49: 111-7.

46 Melnikova VO, Ananthaswamy HN. Cellular and molecular events leading to the development of skin cancer. Mutat Res 2005; 571: 91-106.

47 Garbe C, Buettner PG. Predictors of the use of sunscreen in dermatological patients in Central Europe. Prev Med 2000; 31: 134-9.

48 Glanz K, Buller DB, Saraiya M. Reducing ultraviolet radiation exposure among outdoor workers: state of the evidence and recommendations. Environ Health 2007; 6: 22.

49 LeBlanc WG, Vidal L, Kirsner RS, et al. Reported skin cancer screening of US adult workers. J Am Acad Dermatol 2008; 59: 55-63.

50 Stockfleth E, Kerl H. Guidelines for the management of actinic keratoses. Eur J Dermatol 2006; 16: 599-606.