Arch Dermatol Res (2014) 306:851–856 DOI 10.1007/s00403-014-1503-5

CONCISE COMMUNICATION

A prospective study of cigarette smoking and basal cell carcinoma M. C. B. Hughes • C. M. Olsen • G. M. Williams A. C. Green



Received: 1 June 2014 / Revised: 8 September 2014 / Accepted: 15 September 2014 / Published online: 19 September 2014 Ó Springer-Verlag Berlin Heidelberg 2014

Abstract To investigate the relationship between smoking and primary basal cell carcinoma (BCC), we analyzed data from a 16 year prospective study among randomly selected adults in Nambour, Queensland, Australia. Participants underwent a skin examination in 1992 and took part in an intervention study and follow-up. Information about complexion type and smoking habits including duration and number of cigarettes smoked per day and sun exposure behavior were collected at baseline in 1992, with updates to end of follow-up in 2007. Newly-diagnosed BCCs were ascertained from regional pathology laboratories. Relative risks (RR) of BCC among former and current smokers were estimated using generalized linear models specifying a Poisson distribution with robust error variance and (log) person-years at-risk as offset, adjusting for BCC risk factors. From 1992 to 2007, 281 BCCs were diagnosed in 1,277 participants with available smoking history and no past BCC. Relative to non-smokers, a non-significant inverse association between current smoking and BCC was

M. C. B. Hughes, C. M. Olsen contributed equally to this manuscript and recognized as equal first authors. M. C. B. Hughes (&)  C. M. Olsen  A. C. Green Population Health Department, QIMR Berghofer Medical Research Institute, 300 Herston Rd, Brisbane, QLD 4006, Australia e-mail: [email protected] G. M. Williams School of Population Health, The University of Queensland, Herston Road, Herston, QLD 4006, Australia A. C. Green Institute of Inflammation and Repair, Manchester Academic Health Sciences Centre, University of Manchester, Manchester, UK

seen (RR 0.69; 95 % CI 0.45–1.05) but not for former smokers (RR 1.05; 95 % CI 0.84–1.31). Amongst current smokers, inverse associations with BCC were found in those who smoked for up to 18 years (RR 0.44) but not more and those who smoked up to 15 cigarettes per day but not more. The associations with both current and former smoking varied by degree of sunburn propensity. The modest inverse association between current smoking and BCC is considered unlikely to be causal given lack of clear relation with duration or intensity of smoking. Keywords Basal cell carcinoma  Smoking  Tobacco  Skin cancer  Cohort

Introduction Basal cell carcinoma (BCC) is a major global public health burden due to its very high incidence relative to all other cancers [2, 8, 30]. While the major cause of BCC is solar ultraviolet radiation [13], other lifestyle factors may also be involved [17, 28]. Cigarette smoking may play a role in BCC causation given its moderately high prevalence in most populations and established carcinogenic effects. Several studies, including two recent meta-analyses, have explored the relation between smoking and BCC with inconsistent results. The larger meta-analysis [18] reported no increased risk of BCC with ever-smoking (pooled OR (pOR) 0.95; 95 % CI 0.82–1.09). The second reported a slightly raised BCC risk amongst ever smokers (pOR 1.02; 95 % CI 1.00–1.04) [29]. Neither considered current smokers nor former smokers separately. Longitudinal studies also reported conflicting results. The Nurses’ Health Study reported no increased risk among current, but an increase among former smokers (RR 1.08; 95 % CI

123

852

1.05–1.11); while the Health Professionals Follow-up Study found a significant reduction in BCC risk among current smokers (RR 0.80; 95 % CI 0.73–0.89) and no association with former smokers [29]. Given the heterogeneity of the published literature on this topic, we examined the association between smoking habits and incidence of primary cutaneous BCC in a prospective cohort study. We used smoking at baseline to classify current and former smokers, and fully adjusted our analysis for established risk factors for BCC.

Methods Study participants were drawn from the Nambour Skin Cancer Study initiated in 1986 [11]. Briefly, a random sample of 2,095 residents of Nambour (latitude 26° S) aged 20–69 years, had a skin examination conducted by a dermatologist and completed a questionnaire on personal characteristics, skin phenotype and sun exposure. In 1992, participants were invited to take part in a 5 year field trial of daily sunscreen application and beta-carotene supplementation for skin cancer prevention; 1,621 people consented and underwent a full skin examination by a dermatologist [10, 12]. Eligible participants for this analysis had undergone a skin examination in 1992 and consented to ongoing study participation.

Arch Dermatol Res (2014) 306:851–856

not smoking. Pack-years is a crude measure of smoking dose [27] and was not analyzed. Relative risks (RR) of developing a BCC associated with smoking status were estimated using generalized linear models specifying a Poisson distribution with robust error variance and (log) person-years at-risk as offset [32]. Person-years were computed from the date of skin examination in 1992 to date of death or withdrawal from active participation or 31 December 2007, whichever came first. Age, sex, education, sunscreen and betacarotene treatment allocation and skin color were included a priori in regression models; all other variables (country of birth, baseline occupational sun exposure and post-trial sun exposure) were selected using backward elimination. To investigate possible effect modification for former and current smokers, we stratified analyses by known risk factors and by body site of the first incident BCC. Recent studies have shown that caffeine intake may protect against BCC [7, 22]. Thus, we repeated analyses also adjusting for caffeine intake. We also modeled clinical signs of sun damage at baseline (elastosis of the neck, presence of telangiectasia, freckling on the back and total number of actinic keratoses) in lieu of baseline occupational sun exposure. All analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC, USA).

Data collection

Results

Participants provided information on current and past smoking habits, including duration (total years of smoking) and intensity (average number of cigarettes smoked per day) at baseline (1992). Occupational and recreational sun exposure and usual dietary intake were collected at beginning and end of the field trial. From 1997, sun exposure data were updated twice yearly until 2001 and yearly until 2007. The study outcome was occurrence of histologically confirmed BCC after the baseline skin examination from 1992 to 31 December 2007. From 1992 to 2007, records of BCCs (and other skin cancers) that were biopsied or treated were obtained with participants’ consent from all regional pathology laboratories.

Characteristics of study population

Statistical analyses Smoking status was as reported in 1992 for primary analysis and as at the time of diagnosis of first incident BCC [20] for secondary analysis. Duration of smoking was calculated as the time from age of starting smoking to age of stopping permanently, subtracting reported time spent

123

There were 1,277 participants (mean age 48 yrs) after excluding 202 who had one or more BCCs before the 1992 survey and 142 with missing smoking history. At least one BCC was diagnosed in 281 participants. Of the 996 with no BCC, there were 102 deaths and 72 withdrawals from active questionnaire completion. Participants with a BCC were significantly older, more likely to be born in Australia or New Zealand, have fair skin and report higher levels of post-trial sun exposure (Table 1). Compared with current and former smokers, never smokers were more likely to be women and to be born in Australia or New Zealand, and less likely to report high post-trial sun exposure, to have worked in mainly outdoor occupations and/or to undertake mainly outdoor leisure activities. Current smokers were significantly younger than former smokers, smoked on average more cigarettes per day (mean 19 vs. 17) and for significantly longer duration (mean 25 years vs.17).

Arch Dermatol Res (2014) 306:851–856

853

Table 1 Baseline demographic and personal characteristics of incident basal cell carcinoma cases (1992–2007) and non-cases in the Nambour Skin Cancer Study Characteristics

No BCC n = 996 N (%)

C1 BCC n = 281 N (%)

335 (34)

44 (16)

40–50

282 (28)

68 (24)

50–60

183 (18)

80 (28)

196 (20)

89 (32)

\0.0001

Mean (±SD)

47.0 (12.8)

53.3 (12.3)

\0.0001

Female

575 (58)

153 (54)

Male

421 (42)

128 (46)

Sex

887 (89)

261 (93)

108 (11)

19 (7)

High school or less

487 (49)

128 (46)

Higher education

428 (43)

122 (43)

81 (8)

31 (11)

Fair

521 (52)

164 (58)

Medium

398 (40)

106 (38)

76 (8)

11 (4)

Elsewhere

0.33

0.045

0.57a

Skin color

Olive/dark

0.045

Propensity to sunburn Always burn

182 (18)

59 (21)

Burn then tan

702 (71)

200 (71)

Only tan

111 (11)

22 (8)

Occupation Professional Non-professional

271 (27)

79 (28)

715 (73)

200 (72)

0.20

0.78

Occupational sun exposure Mainly outdoors

182 (18)

55 (20)

Indoors and outdoors

354 (36)

114 (41)

Mainly indoors

459 (46)

112 (40)

Mainly outdoors

400 (40)

128 (46)

Indoors and outdoors

449 (45)

112 (40)

Mainly indoors

145 (15)

41 (15)

0.16

Leisure-time sun exposure

0.23

Lifetime number of sunburns B1 sunburn

288 (29)

81 (29)

2–5

440 (44)

119 (42)

[5

267 (27)

81 (29)

Hardly ever \50 % of the time

122 (12) 576 (58)

21 (7) 172 (61)

[50 % of the time

251 (25)

84 (30)

47 (5)

4 (1)

Missing

Discretionary sunscreen

511 (51)

131 (47)

Daily sunscreen

485 (49)

150 (53)

P value

0.17

a

Placebo

488 (49)

134 (48)

Daily beta-carotene

508 (51)

147 (52)

0.70

Not including missing category

After adjustment for confounding factors, there was a nonsignificant inverse association between smoking and BCC among current smokers (RR 0.69; 95 % CI 0.45–1.05) but not former smokers (RR 1.05; 95 % CI 0.84–1.31) relative to never smokers (Table 2). An inverse association was apparent for those who smoked for up to 18 years (RR 0.44; 95 % CI 0.17–1.16), but not in those who had smoked for longer (Table 2). BCC risk amongst former smokers did not vary materially with duration of smoking. Amongst current smokers, smoking up to 15 cigarettes per day was inversely associated with BCC (OR 0.49; 95 % CI 0.24–1.02), but smoking more than this was not (Table 2). Analyses of current smoking and BCC stratified by body site of BCC were based on small numbers but showed strong inverse associations for BCC on the trunk (RR 0.34, 95 % CI 0.13–0.91) and upper limbs (RR 0.27; 95 % CI: 0.04-2.04) but not with BCC of the head and neck (RR 0.78; 95 % CI 0.44, 1.40) or lower limbs (RR 1.60; 95 % CI 0.56–4.59). Risk of BCC with current smoking did not vary substantially by sex, age, post-trial or occupational sun exposure, presence of actinic keratoses, skin color, trial sunscreen treatment or occupation type, but did differ by propensity to sunburn (Fig. 1a). Former smoking was inversely associated with BCC in those least prone to sunburn or olive-skinned (Fig. 1b). Using objective signs of sun damage in lieu of occupational sun exposure, adjusting for level of caffeine intake, or taking smoking status at the time of first incident BCC did not materially change results (not shown).

0.78

Post-trial sun exposure

Sunscreen trial allocation

C1 BCC n = 281 N (%)

Smoking and BCC

Education

Missing

No BCC n = 996 N (%)

Beta-carotene trial allocation

60?

Country of birth Australia or New Zealand

Characteristics

P value

Age at baseline \40 years

Table 1 continued

Discussion 0.0421

We prospectively examined the association between smoking habit and BCC incidence in a population-based cohort, taking full account of established risk factors. A non-significant inverse association was found between

123

854 Table 2 Baseline smoking status and incident cutaneous BCC from1992 to 2007, stratified by duration and intensity of smoking

Arch Dermatol Res (2014) 306:851–856

No BCC Count (%)

C1BCC Count (%)

Crude RR (95 %CI)

Adjusteda RR (95 %CI)

Never smoker

551 (55)

164 (58)

1.00

1.00

Former smoker

289 (29)

95 (34)

1.08 (0.87, 1.35)

1.05 (0.84, 1.31)

Current smoker

156 (16)

22 (8)

0.57 (0.38, 0.86)

0.69 (0.45, 1.05)

551 (55)

164 (58)

1.00

1.00

\=18 years

168 (17)

50 (18)

0.99 (0.75, 1.31)

1.13 (0.86, 1.49)

18?

121 (12)

45 (16)

1.20 (0.90, 1.59)

0.97 (0.72, 1.29)

63 (6) 93 (9)

4 (1) 18 (6)

0.26 (0.10, 0.68) 0.78 (0.51, 1.21)

0.44 (0.17, 1.16) 0.78 (0.50, 1.23)

551 (55)

164 (58)

1.00

1.00

\=15 cigarettes per day

162 (16)

46 (16)

0.95 (0.72, 1.27)

0.96 (0.73, 1.27)

15? cigarettes per day

127 (13)

49 (17)

1.24 (0.94, 1.62)

1.16 (0.87, 1.53)

Smoking status at baseline Smoking status

Duration of smoking Never smoker Former smoker

Current smoker \=18 years 18? Intensity of smoking Never smoker Former smoker a

Adjusted for age, sex, country of birth, skin color, trial sunscreen treatment, trial betacarotene treatment, education, occupational sun exposure at baseline and post-trial sun exposure

Current smoker \=15 cigarettes per day

73 (7)

7 (2)

0.43 (0.21, 0.87)

0.49 (0.24, 1.02)

15? cigarettes per day

83 (8)

15 (5)

0.69 (0.43, 1.11)

0.85 (0.52, 1.38)

current smoking and BCC that was stronger after short rather than after long duration of smoking and stronger in light compared with heavy smokers. There was no association among former smokers overall. The associations of both current and former smoking with BCC varied (in a reciprocal fashion) with degree of sunburn propensity. Our results are somewhat in agreement with other cohort data. The Health Professionals Follow-up Study reported an inverse association of BCC with current smoking, and no association with former smoking [29]. However, the Nurses’ Health Study showed no association of current smoking with BCC and a small increase in risk with former smoking [29]. The Radiologic Technologists cohort [9] reported a modest inverse association between current smoking and BCC of borderline significance and a weak positive association with past smoking. Another US cohort study [16] found that amongst people with past keratinocyte cancer, BCC was inversely associated with heavy smoking, but not with smoking duration. Case–control data are even more inconsistent. One study [4] reported an inverse association only for superficial multifocal BCC, and a twin study reported an inverse association for men and an increase for women [21], while a third study found a non-significant increased risk in current smokers [3]. Two meta-analyses reported conflicting results due to the inclusion of different case–control studies in each meta-analysis [18, 29]. Both compared ‘ever smokers’ to ‘never smokers’ only.

123

The perplexing finding of an inverse association between smoking and BCC has also been reported for melanoma [5]. In contrast, cohort studies have reported a positive association of smoking with cutaneous squamous cell carcinoma (SCC) [29]. Our own prospective data from the Nambour study showed no effect of current or former smoking on SCC [20]. Tobacco smoke suppresses immunologic functions [15], resulting in a loss of immune surveillance, an important element in skin carcinogenesis. Whilst it is plausible that smoking affects the development of BCC and SCC very differently, other explanations such as confounding by socio-economic status, obesity and/or lack of physical activity, or caffeine intake appear more likely, although our extensive exploratory analyses failed to find evidence of this. The lack of association for high cumulative amounts smoked by current smokers suggests a non-causal relationship. At baseline, 75 % of former smokers had quit smoking 25 years ago or more and so resembled never-smokers, explaining the lack of overall association between former smokers and BCC. Smoking is associated with decreased risks of inflammatory bowel [19] and Parkinson’s disease [24] but the only malignancy consistently inversely related to cigarette smoking is endometrial cancer [1, 26], believed reflecting lower levels of circulating estrone and estradiol amongst smokers [14]. With some exceptions [25], hormonal factors are not known to be associated with BCC. Smoking may also interact with genetic factors to alter BCC risk: either

Arch Dermatol Res (2014) 306:851–856

855

Fig. 1 Results of stratified analyses on the association between a current smoking and b former smoking at baseline and risk of BCC

pigment-associated genes, as suggested by the effect modification by sunburn propensity we observed, or with loci not linked to pigmentation that appear to confer BCC risk [23, 31]. Strengths of this study include its prospective design, the availability of detailed information about potential confounders, complete follow-up of 77 % of the study population and near-complete ascertainment of histologicallyconfirmed BCCs. We thus consider misclassification of BCC status, unlikely. Limitations include reliance on recalled sun exposure measures that may have resulted in misclassification [6]. However, our sensitivity analyses replacing sun exposure variables with clinical markers of cumulative sun damage did not change results materially. Smoking history is also subject to recall bias, and the use of baseline smoking status may have led to a degree of misclassification. Change in smoking status was recorded in follow-up surveys and was observed in only 111 (9 %) study participants however, and, results using recent smoking status were similar to results using baseline smoking status, demonstrating that exposure misclassification is likely to be small. In conclusion, our data show a modest inverse association between current smoking and BCC but no association with former smoking. Both results varied by degree of sunburn propensity and site of BCC. No clear relation was seen between duration of smoking or amount smoked and

BCC development however, so we consider this association unlikely to be causal. These findings warrant further exploration in large prospective studies. Acknowledgments MCB Hughes, CM Olsen were supported by the National Health and Medical Research Council of Australia (NHMRC) Program Grant (No. 552429). AC Green was supported in part by a fellowship from the Medical Research Council (grant number 89912). Conflict of interest of interest.

The authors declare that they have no conflict

References 1. Baron JA (1996) Beneficial effects of nicotine and cigarette smoking: the real, the possible and the spurious. Br Med Bull 52:58–73 2. Chen JG, Fleischer AB Jr, Smith ED, Kancler C, Goldman ND, Williford PM, Feldman SR (2001) Cost of nonmelanoma skin cancer treatment in the United States. Dermatol Surg 27:1035–1038 3. Corona R, Dogliotti E, D’Errico M, Sera F, Iavarone I, Baliva G, Chinni LM, Gobello T, Mazzanti C, Puddu P, Pasquini P (2001) Risk factors for basal cell carcinoma in a Mediterranean population: role of recreational sun exposure early in life. Arch Dermatol 137:1162–1168 4. De Hertog SA, Wensveen CA, Bastiaens MT, Kielich CJ, Berkhout MJ, Westendorp RG, Vermeer BJ, Bouwes Bavinck JN (2001) Relation between smoking and skin cancer. J Clin Oncol 19:231–238

123

856 5. DeLancey JO, Hannan LM, Gapstur SM, Thun MJ (2011) Cigarette smoking and the risk of incident and fatal melanoma in a large prospective cohort study. Cancer Causes Control 22:937–942 6. English DR, Armstrong BK, Kricker A (1998) Reproducibility of reported measurements of sun exposure in a case-control study. Cancer Epidemiol Biomarkers Prev 7:857–863 7. Ferrucci LM, Cartmel B, Molinaro AM, Leffell DJ, Bale AE, Mayne ST (2014) Tea, coffee, and caffeine and early-onset basal cell carcinoma in a case-control study. Eur J Cancer Prev 23:296–302 8. Fransen M, Karahalios A, Sharma N, English DR, Giles GG, Sinclair RD (2012) Non-melanoma skin cancer in Australia. Med J Aust 197:565–568 9. Freedman DM, Sigurdson A, Doody MM, Mabuchi K, Linet MS (2003) Risk of basal cell carcinoma in relation to alcohol intake and smoking. Cancer Epidemiol Biomarkers Prev 12:1540–1543 10. Green A, Battistutta D, Hart V, Leslie D, Marks G, Williams G, Gaffney P, Parsons P, Hirst L, Frost C et al (1994) The Nambour skin cancer and actinic eye disease prevention trial: design and baseline characteristics of participants. Control Clin Trials 15:512–522 11. Green A, Beardmore G, Hart V, Leslie D, Marks R, Staines D (1988) Skin cancer in a Queensland population. J Am Acad Dermatol 19:1045–1052 12. Green A, Williams G, Neale R, Hart V, Leslie D, Parsons P, Marks GC, Gaffney P, Battistutta D, Frost C, Lang C, Russell A (1999) Daily sunscreen application and betacarotene supplementation in prevention of basal-cell and squamous-cell carcinomas of the skin: a randomised controlled trial. Lancet 354:723–729 13. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans (2012) Radiation: solar and UV radiation. IARC Monogr Eval Carcinog Risks Hum 100(Pt D):35–101 14. Jensen J, Christiansen C, Rodbro P (1985) Cigarette smoking, serum estrogens, and bone loss during hormone-replacement therapy early after menopause. N Engl J Med 313:973–975 15. Johnson JD, Houchens DP, Kluwe WM, Craig DK, Fisher GL (1990) Effects of mainstream and environmental tobacco smoke on the immune system in animals and humans: a review. Crit Rev Toxicol 20:369–395 16. Karagas MR, Stukel TA, Greenberg ER, Baron JA, Mott LA, Stern RS (1992) Risk of subsequent basal cell carcinoma and squamous cell carcinoma of the skin among patients with prior skin cancer. Skin Cancer Prevention Study Group. J Am Med Assoc 267:3305–3310 17. Kaskel P, Lange U, Sander S, Huber MA, Utikal J, Leiter U, Krahn G, Meurer M, Kron M (2014) Ultraviolet exposure and risk of melanoma and basal cell carcinoma in Ulm and Dresden. J Eur Academy Dermatol Venereol JEADV, Germany. doi:10. 1111/jdv.12488 18. Leonardi-Bee J, Ellison T, Bath-Hextall F (2012) Smoking and the risk of nonmelanoma skin cancer: systematic review and meta-analysis. Arch Dermatol 148:939–946 19. Mahid SS, Minor KS, Soto RE, Hornung CA, Galandiuk S (2006) Smoking and inflammatory bowel disease: a meta-analysis. Mayo Clin Proc 81:1462–1471 20. McBride P, Olsen CM, Green AC (2011) Tobacco smoking and cutaneous squamous cell carcinoma: a 16 year longitudinal

123

Arch Dermatol Res (2014) 306:851–856

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

population-based study. Cancer Epidemiol Biomarkers Prev 20:1778–1783 Milan T, Verkasalo PK, Kaprio J, Koskenvuo M (2003) Lifestyle differences in twin pairs discordant for basal cell carcinoma of the skin. Br J Dermatol 149:115–123 Miura K, Hughes MC, Green AC, van der Pols JC (2014) Caffeine intake and risk of basal cell and squamous cell carcinomas of the skin in an 11 year prospective study. Eur J Nutr 53:511–520 Nan H, Xu M, Kraft P, Qureshi AA, Chen C, Guo Q, Hu FB, Curhan G, Amos CI, Wang LE, Lee JE, Wei Q, Hunter DJ, Han J (2011) Genome-wide association study identifies novel alleles associated with risk of cutaneous basal cell carcinoma and squamous cell carcinoma. Hum Mol Genet 20:3718–3724 Noyce AJ, Bestwick JP, Silveira-Moriyama L, Hawkes CH, Giovannoni G, Lees AJ, Schrag A (2012) Meta-analysis of early nonmotor features and risk factors for Parkinson disease. Ann Neurol 72:893–901 Olsen CM, Hughes MC, Pandeya N, Green AC (2006) Anthropometric measures in relation to basal cell carcinoma: a longitudinal study. BMC Cancer 6:82 Pfeiffer RM, Park Y, Kreimer AR, Lacey JV Jr, Pee D, Greenlee RT, Buys SS, Hollenbeck A, Rosner B, Gail MH, Hartge P (2013) Risk prediction for breast, endometrial, and ovarian cancer in white women aged 50 years or older: derivation and validation from population-based cohort studies. PLoS Med 10:e1001492 Pirie K, R Peto, Reeves GK, Green J, Beral V, Million Women Study C (2013) The 21st century hazards of smoking and benefits of stopping: a prospective study of one million women in the UK. Lancet 381:133–141 Richmond-Sinclair NM, Pandeya N, Ware RS, Neale RE, Williams GM, van der Pols JC, Green AC (2009) Incidence of basal cell carcinoma multiplicity and detailed anatomic distribution: longitudinal study of an Australian population. J Invest Dermatol 129:323–328 Song F, Qureshi AA, Gao X, Li T, Han J (2012) Smoking and risk of skin cancer: a prospective analysis and a meta-analysis. Int J Epidemiol 41:1694–1705 Souza RJ, Mattedi AP, Correa MP, Rezende ML, Ferreira AC (2011) An estimate of the cost of treating non-melanoma skin cancer in the state of Sao Paulo, Brazil. Anais brasileiros de dermatologia 86:657–662 Stacey SN, Gudbjartsson DF, Sulem P, Bergthorsson JT, Kumar R, Thorleifsson G, Sigurdsson A, Jakobsdottir M, Sigurgeirsson B, Benediktsdottir KR, Thorisdottir K, Ragnarsson R, Scherer D, Rudnai P, Gurzau E, Koppova K, Hoiom V, Botella-Estrada R, Soriano V, Juberias P, Grasa M, Carapeto FJ, Tabuenca P, Gilaberte Y, Gudmundsson J, Thorlacius S, Helgason A, Thorlacius T, Jonasdottir A, Blondal T, Gudjonsson SA, Jonsson GF, Saemundsdottir J, Kristjansson K, Bjornsdottir G, Sveinsdottir SG, Mouy M, Geller F, Nagore E, Mayordomo JI, Hansson J, Rafnar T, Kong A, Olafsson JH, Thorsteinsdottir U, Stefansson K (2008) Common variants on 1p36 and 1q42 are associated with cutaneous basal cell carcinoma but not with melanoma or pigmentation traits. Nat Genet 40:1313–1318 Zou G (2004) A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 159:702–706

A prospective study of cigarette smoking and basal cell carcinoma.

To investigate the relationship between smoking and primary basal cell carcinoma (BCC), we analyzed data from a 16 year prospective study among random...
193KB Sizes 1 Downloads 6 Views