- Email: [email protected]
Skin cancer in Australian heart transplant recipients
Skin cancer in Australian heart transplant recipients Colin S. Ong, MBBS,b Anne M. Keogh, MD,a Steven Kossard, FACD,b Peter S. Macdonald, PhD,a and Phillip M. Spratt, FRCSa Darlinghurst, Australia Background: Cutaneous malignancy is a major cause of morbidity in organ transplant recipients. Objective: Our purpose was to report on skin cancer in Australian heart transplant recipients with analysis of HLA factors. Methods: We reviewed histologically proven skin cancers in the first 455 patients undergoing organ transplantation in Sydney, Australia. Results: The cumulative incidence of skin cancer was 31% at 5 years and 43% at 10 years with a squamous cell carcinoma/basal cell carcinoma ratio of 3:1. Caucasian origin, increasing age at transplantation, and duration of follow-up were significantly associated with skin cancer. Skin cancer accounted for 27% of 41 deaths occurring after the fourth year. Recipient HLA-DR homozygosity was associated with skin cancer overall, whereas HLA-DR7 was a protective factor in skin cancer overall, squamous cell carcinoma, and Bowen’s disease. HLA-A1 and HLA-A11 were significant protective factors in Bowen’s disease. Conclusion: Skin cancer is a major cause of morbidity and long-term mortality in heart transplant patients. (J Am Acad Dermatol 1999;40:27-34.)
Skin cancer is the most common malignancy affecting organ transplant recipients.1,2 Immunosuppressed patients are more predisposed to skin malignancy, which often occurs at an earlier age1-5 and is often multifocal and more aggressive in nature than in nonimmunosuppressed persons.1,2,4,6 Cutaneous malignancy is a significant cause of morbidity and can also be a major cause of malignancy-related mortality. Apart from druginduced immunosuppression and previous sun exposure, human papillomavirus (HPV) and immunogenetic factors (HLA haplotype and patient-donor mismatching) have been reported to play a role in the development of skin cancer. Heart transplant recipients experience a greater rate of skin cancer and other malignancies than kidney transplant recipients.7,8 This has been attributed to the slightly higher doses of immunoFrom the Cardiopulmonary Transplant Unit, St Vincent’s Hospital,a and Skin and Cancer Foundation Australia.b Accepted for publication Sept 15, 1998. Reprint requests: Associate Professor Steven Kossard, Skin and Cancer Foundation, 277 Bourke St, Darlinghurst, NSW, 2010, Australia. Copyright © 1999 by the American Academy of Dermatology, Inc. 0190-9622/99/$8.00 + 0 16/1/94446
suppression agents used in heart transplantation compared with other solid organ transplants. Australia has one of the highest rates of skin malignancy in the world. The rate of nonmelanoma skin cancer for the period 1985-1992 in the most northern state of the country (Queensland) was noted for basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) to be 2074 and 1579 per 100,000 population per year, respectively. SCC occurred at half the rate of BCC among men and at approximately one third the rate among women.9 In 1990, the rate of nonmelanoma skin cancer in northern Australia was up to 4 times that in southern Australia,10 whereas the incidence rate of melanoma for 1989 in Queensland was twice that in southernmost Australia.11 As expected, the rate of skin cancer in kidney transplant recipients in Australia is also one of the highest because previous sun exposure appears to be a major factor in the development of skin cancers. For kidney transplant recipients in Queensland, the rate was noted to be 45% after 11 years, rising to 70% after 20 years of immunosuppression.12 Other countries with high sun exposure such as Spain also have a high cumulative risk for skin cancer of 43.8% after 7 years after heart 27
Journal of the American Academy of Dermatology January 1999
28 Ong et al Table I. Skin lesions and patients affected Total
SCC BCC KA Melanoma Bowen’s disease Total
Caucasian
Non-Caucasian
Patients affected
No. of lesions
No. of patients
No. of lesions
No. of patients
No. of lesions
113 92 19 7 79 152
849 285 28 7 263 1436*
111 91 19 7 75 148
844 278 28 7 256 1417*
2 1 0 0 4 4
5 7 0 0 7 19
KA, Keratoacanthoma. *Includes 3 atypical fibroxanthomas and one Merkel cell tumor.
Table II. Distribution of lesions Location
No. of lesions (%)
Head and neck Trunk Upper limb Lower limb
929 (65) 207 (14) 183 (13) 117 (8)
transplantation.13 There have been no reports to date on the rates of skin cancer in Australian heart transplant recipients. In this study, we examine the rate of skin cancer in Australian heart transplant recipients and its relation to various factors, including HLA antigens. MATERIAL AND METHODS The heart-lung transplant unit at St Vincent’s Hospital in Sydney, Australia performed 620 heart transplants between 1984 and January 1998. This retrospective study examines outcomes in the first 455 patients undergoing transplantation. Information on skin cancers was gathered from several sources, including patients, medical practitioners, medical records, and hospitals in the compilation of the Transplant Unit database. Only histologically confirmed malignancies were included in the analysis. Descriptive statistics were performed on Microsoft Excel 5.0 and inferential statistics performed on Statview 4.5 (Abacus software). The SAS statistical software package (SAS Institute Inc) was used for univariate and multivariate logistic regression, including the stepwise selection procedure. Forty-five patients who died within 6 months after transplant and 10 patients for whom information was incomplete were excluded from the study. As a result 400 patients were studied who experienced a total of 1436 skin lesions (mainly SCC, BCC, keratoacanthoma, Bowen’s disease, and melanoma). The age of
patients ranged from 6.6 to 67 years at transplantation (median, 47.9 years). Fourteen patients were younger than 16 years of age at transplantation. The median follow-up of the 400 patients was 5.5 years (minimum, 0.5 years; maximum 13 years) taken from transplantation to March 1997 or date of death. Two hundred fifteen patients had a follow-up period of more than 5 years. Of the study population, 347 (87%) were of Caucasian background, 31 (8%) of Mediterranean origin, 16 (4%) of Asian origin, and 6 (1.5%) of Aboriginal descent. Immunosuppressive protocols comprised therapy with azathioprine and cyclosporine with or without prednisolone. Rejection episodes of grade greater than 3a in the first 4 weeks after transplantation were treated with intravenous pulsed methylprednisolone and beyond 4 weeks with an oral prednisolone taper. More resistant rejection was treated variously with OKT-3, antithymocyte globulin, total lymphoid irradiation or more recently with conversion from cyclosporine to FK 506 (tacrolimus) or from azathioprine to mycophenolate mofetil. RESULTS
Lesions developed in 152 of 400 patients (38%). A total of 1436 lesions (Table I) were noted with an SCC/BCC ratio of 3:1; a larger proportion of SCC in male than in female patients was found (3.2:1 vs 1.3:1). The majority of lesions occurred on the head and neck, with the least on the lower limb (Table II). The crude cumulative incidence rates of skin cancer were 8.5%, 21%, 31%, and 43% after 1, 3, 5, and 10 years after transplantation, respectively (Kaplan-Meier estimate). Four male patients had more than 50 skin cancers. Metastases developed in 9 male patients with SCC and 4 with melanoma. Eleven deaths from skin cancer (6 from SCC, 4 from melanoma, 1 from Merkel cell carcinoma) accounted for 27% of
Journal of the American Academy of Dermatology Volume 40, Number 1
Ong et al 29
Table III. Multivariate logistic regression models: Odds ratio (95% confidence interval) Factor
Caucasian origin Age (y) at transplantation* Duration (y) of follow-up* HLA-DR7 HLA-DR homozygosity HLA-A1 HLA-A11 Residence in northern Australia
Cancer overall
BCC
SCC
Bowen’s disease
15.1 (4.8-47.3) 1.6 (1.4-1.9) 5.0 (3.2-7.9) 0.4 (0.2-0.7) 2.0 (1.1-3.8) — — —
20.7 (2.8-154.6) 1.5 (1.3-1.8) 2.7 (1.7-4.2) — — — — 2.3 (1.0-5.4)
16.3 (3.6-73.4) 1.7 (1.4-1.9) 6.0 (3.7-9.8) 0.5 (0.2-0.9) — — — —
4.1 (1.3-13.2) 1.6 (1.3-1.9) 6.5 (3.8-11.3) 0.4 (0.2-0.8) — 0.5 (0.3-0.9) 0.4 (0.1-1.0) —
*For being 5 years older or having had a further 5 years of follow-up.
deaths occurring after the fourth year after transplantation. Smoking was not significantly related to the occurrence of skin or lip cancers. The overall incidence of SCC was 37,949 per 100,000 (849/2237.2) posttransplant person-years; that of BCC, 12,739 per 100,000 (285/2237.2) posttransplant person-years; that of Bowen’s disease, 11,755 per 100,000 (263/2237.2) personyears; and that of melanoma, 312 per 100,000 (7/2237.2) person-years. Overall, BCC occurred earlier than SCC after transplantation (4.3 ± 2.6 vs 6.2 ± 2.5 years [mean ± standard deviation]; P < .0001, 2-sample t test). The number of skin cancers was significantly correlated with both age at transplantation (P < .0001, τ 0.168, Kendall rank correlation corrected for ties) and duration of follow-up (P < .0001, τ 0.246). Both were significantly correlated with the number of cancers at different sites and to subtypes of cancer (all P values < .03 except melanoma). Patients with skin cancer were significantly older at transplantation than those with no skin cancer (50.0 vs 45.8 years, median; P < .0001, Mann-Whitney U test). The median age of patients with skin cancer at diagnosis of their first lesion was 54.0 years. On univariate analysis, Caucasian origin (compared with non-Caucasian) was associated with skin cancer: odds ratio 18.5 (95% confidence interval, 2.5-135.6) for BCC occurrence, 12.0 (2.9-50.1) for SCC, and 3.4 (1.2-9.7) for Bowen’s disease. On univariate logistic regression, skin cancer occurrence was more frequent among male than female patients (odds ratio 1.9; 95% confidence interval 1.1-3.3), with all types being more common among male patients but not significantly so. Male patients had a significantly greater
number of SCC (P = .03, Mann-Whitney U test) and total lesions than female patients (P = .02), but only among Caucasian patients. Male and female patients had similar follow-up periods, but male patients were significantly older at transplantation (P = .002) and experienced fewer treated rejection episodes (P = .01). Residing in northern Australia (Northern Territory and Queensland) just before transplantation was significantly associated with all types of skin cancer on univariate analysis (odds ratios, 2.0-2.7). Stepwise selection was used to determine a multivariate logistic regression model for each type of cancer; only variables meeting the .05 significance level were entered into the model. The odds ratios of having a particular type of cancer or cancer overall are listed in Table III. Although skin cancer was associated with being male and residing in northern Australia, these factors did not remain significant in multivariate analysis except for the latter in BCC. We did not find any other HLA antigens, the occurrence of HLA antigen mismatching, or the number of treated rejection episodes to be significantly associated with skin cancer occurrence. Polytomous regression did not provide additional findings. DISCUSSION
Our study confirms the high incidence of skin cancer in organ transplant recipients, as found in the study on Queensland kidney transplant recipients.12 Our high incidence may be partly due to inclusion of lesions other than BCC and SCC that were not included in that study. However, it is more likely that we may have underestimated the incidence of skin cancer because some lesions, in particular BCCs, would have been treated by non-
Journal of the American Academy of Dermatology January 1999
30 Ong et al Table IV. Studies of skin cancer in heart transplant recipients Authors
No. of patients
Euvrard et al7* (France)
150
Espana et al13 (Spain)
92
Age (y) at transplant
55.7 ± 6.2
Follow-up (y)
Findings
Not stated†
136 skin lesions in 41 patients (27%); SCC/BCC ratio: 1.08:1 26 skin cancers in 14 patients (15.2%); SCC/BCC ratio: 1.5:1 96 lesions total; 12.9% had malignant lesions; SCC/BCC ratio: 1:1.08 20 lesions in 20 patients (6.5%); SCC/BCC ratio: 1:1
49.5 (0.17-69)
3.6 (0.1-9.5)
Jensen et al14 (Norway)
140
47.7 (2-63)
5 (1-10)
Mihalov et al15* (US)
307
45.1 (6-71)
2.8
*Other solid organ transplant recipients studied have not been included in this table. †This study examined 150 patients referred for dermatologic problems over a 5-year period.
surgical means and thus would not have been included in the study. The estimates from our study may be lower than some other studies because we have included patients of nonCaucasian background (13.2%) and a proportion of our patients of Angloceltic and European origin had spent their early years in Northern Europe, which would have tended to reduce their cumulative sun exposure. Skin cancers developed in 2 of 6 Aboriginal patients and in 2 of 16 patients of Mediterranean origin. Logistic regression showed that patients of Caucasian origin and those residing in northern Australia just before transplantation had a higher frequency of skin cancers, although the latter remained significant only for BCC in multivariate analysis. The finding of a 3:1 SCC/BCC ratio is consistent with previous studies (Table IV) in which a reversal in the SCC/BCC ratio was noted when compared with the general population.1,7,13 In kidney transplant recipients, the SCC/BCC ratio has been noted to be up to 16:1.16 The greater SCC/BCC ratio for male patients as compared with female patients (3.2:1 vs 1.3:1) can be explained by the greater number of lesions male patients have after transplantation. On multivariate analysis, although there was a strong trend for some skin cancers, sex did not remain a significant factor, indicating that other factors may be more important. Mihalov et al15 found that SCC occurred later than BCC in heart transplant patients, whereas the opposite was true of kidney transplant patients. In addition, BCC occurred earlier and SCC later in heart than in kidney transplant patients. In contrast, Euvrard et al7 found that skin lesions occurred earlier in heart transplant patients, but
BCC did comprise a higher proportion of lesions in their heart transplant patients. In our study, when all SCCs and BCCs are considered, the interval to the development of SCC was significantly longer than for BCC (6.2 ± 2.5 vs 4.3 ± 2.6 years, mean ± SD; P < .0001, 2-sample t test). It appears that BCCs occur earlier after transplantation and SCCs develop and predominate after a longer period of immunosuppression. The correlation of the number of skin cancers with age at transplantation and the duration of follow-up, as well as the odds ratios reported for their occurrence, is consistent with the hypothesis that exposure to sunlight before transplantation and duration of exposure to immunosuppression are major risk factors. Our study also reinforces the concept that immunosuppressed patients have skin cancers at an earlier age. Patients had a median age of 54 years at diagnosis of their first histologically confirmed skin cancer, whereas in the general population the average age of patients with either SCC or BCC is 10 years older,17-19 although those living closer to the equator develop lesions at a considerably younger age.17 There were no cases of Kaposi’s sarcoma in our study, in contrast to the Australian kidney transplant population2 and transplant recipients overseas for whom rates up to 5.3% in Saudi Arabia have been reported.1,20 This can be explained in part by the ethnicity of our transplant population of which 92% were of non-Mediterranean origin and thus not in a high-risk group for Kaposi’s sarcoma. Merkel cell carcinoma developed in 1 patient, atypical fibroxanthoma in 3 patients, leiomyosarcoma in 1, and in 2 patients, lymphoproliferative disease presented initially as skin lesions.
Journal of the American Academy of Dermatology Volume 40, Number 1
In our group, 11 patients died of skin cancer, accounting for 27% of deaths occurring after the fourth year post transplant. Seven patients had melanomas; metastases developed in 4 of these patients and they died of melanoma. In one study, the majority of melanomas occurring after kidney transplantation arose in precursor nevi,21 indicating the need to monitor nevi closely after transplantation. Metastases developed in 9 patients with SCC; 6 died of metastatic SCC. The patient with Merkel cell carcinoma died of metastatic disease. Invasion of the parotid gland developed in 4 patients with SCC and invasion of the external ear requiring radiotherapy developed in 2 patients, highlighting the role of immunosuppression in predisposing patients to more aggressive cancers. The risk of metastases in SCC is estimated to be 1% to 2% in the general population,22 although a rate of 5.8% was found in a hospital-based study.23 Cutaneous SCC secondary to inflammatory and degenerative processes (ostemyelotic foci, burns, scars of lupus vulgaris) developed metastases in 10% to 30% of such cases. Eleven percent of labial SCC metastasize in contrast to 3.3% of skin SCC.24 Apart from origin, anatomic site, immunosuppression, histologic features (such as tumor grade and perineural invasion), size and depth of the tumor, and previous treatment also influence the risk of metastasis.3 In our study, metastases subsequently developed in 8% of patients with SCC, contributed to a large extent by the occurrence of multiple lesions in patients. SCCs, which are the nonmelanoma skin cancer most likely to metastasize, are also the most poorly diagnosed clinically.22 Nixon, Dorevitch, and Marks22 recommend biopsy as a routine to confirm the clinical diagnosis, particularly if the tumor’s histologic type will influence the mode of its removal. The majority of skin cancers occurring on the head and neck supports the role of sun exposure in causation. UV light radiation is not only a carcinogen but can induce tolerance or immunosuppression to tumor antigens by interfering with Langerhans cell antigen presentation.25 Seventy percent of 152 patients with skin cancer on whom a smoking history was available were ex-smokers, suggesting that smoking may contribute to skin cancer as it does in labial, lung, and cervical cancer. Smoking is a risk factor for SCC,26-28 and tobacco smoke condensate can induce SCC in rabbit and mouse skin.29 Smokers with melanoma are
Ong et al 31 more likely to have metastases on initial presentation and have a poorer prognosis than nonsmokers.30 Tobacco smoke condensate can impair function of Langerhans cells31 and affect other parts of the immune system as well.32,33 Both sun exposure and smoking are risk factors for dysplastic and malignant lip lesions in kidney transplant recipients.34 Although we did not find an association between smoking and skin or lip cancer, our results are not conclusive because we have not quantified the extent of smoking by individual patients. Many studies have examined immunogenetic factors, particularly HLA haplotypes in skin cancer. In normal nonimmunosuppressed patients, HLA-DR4 was found to be significantly less common in patients with multiple BCCs, whereas HLA-DR1 was more common.35-37 In transplant recipients, HLA-A11 was found to be protective against skin cancer38 in The Netherlands. However, this was not found in kidney transplant recipients in Southern Australia39 and was associated with skin cancer in northern Australia.40 These differences are possibly related to the relative importance of human papillomavirus or sun exposure in skin cancer development.40 Other antigens found to be associated with skin cancer in kidney transplant recipients include HLADQw2,37 HLA-DR7,39,40 HLA-A3,38 and HLAB27.38,39 Varied findings of Bouwes Bavinck et al38 and Czarnecki et al35,36,39 may be due to the effect of geographic location on the incidence of skin cancer in the transplant and general population. However, both found that HLA-B2738,39 was a strong risk factor for skin cancer in the kidney transplant population. In addition, Bouwes Bavinck et al41 have found that HLA-B mismatching and HLA-DR homozygosity are significantly associated with an increased risk of SCC. In our study, multivariate logistic regression models arising from stepwise selection show that the occurrence of skin cancer is associated with HLA-DR homozygosity. This is consistent with a study of kidney transplant recipients in The Netherlands41 but was not found in northern Australia.40 Our findings that absence of HLADR7 is associated with the occurrence of skin cancer overall, SCC, and Bowen’s disease is in contrast to findings of previous studies.39,40 However, in one of those studies the finding did not reach significance40 and in the other39 the control group
32 Ong et al was a healthy population, rather than kidney transplant recipients. It should be noted that HLA-DR7 is associated with having multiple BCCs in the nontransplant population.42 We did not find a significant relation between BCC occurrence and presence or absence of HLA-DR7. Our finding that HLA-11 is a protective factor in Bowen’s disease is consistent with that of other studies.38 HLA-A1 as a protective factor for Bowen’s disease has not been previously reported. Although our findings are consistent with HLA antigens affecting the immune response to tumor and/or papillomavirus antigens, it is clear that the association of skin cancer and particular HLA antigens is complex because multiple other factors play a role in skin cancer development. We found no relation between the occurrence of skin cancer and the number of treated rejection episodes on univariate and multivariate analysis. We have not investigated the relation of skin cancer to individual immunosuppressive drugs. Any such analysis should consider drug interactions in assessing the level of immunosuppression. Ketoconazole is often used to potentiate the effect of cyclosporine. Diltiazem, which is used to prevent transplant coronary artery disease,43 affects cyclosporine blood levels,44 reducing dosage requirements by one third45 and may inhibit the proliferation of T lymphocytes.46 Early studies suggested azathioprine had a particular risk of inducing keratotic lesions and skin cancer47,48 possibly related to its active metabolite 6-thioguanine, which is a tumor initiator and promotor.48 Lennard et al49 found the level of this metabolite to be associated with the presence of dysplastic lesions. Alternatively, it was thought that metabolites of azathioprine, the imidazole compounds, may sensitize the skin to sunlight, thereby increasing the risk of skin cancer.50 It was then proposed that cyclosporine should be used instead of azathioprine given the perceived risk with azathioprine. On the other hand, cyclosporine may promote the progression of UV radiation–induced skin cancers.51 However, it has been shown that there is no significant difference in skin cancer incidence in kidney transplant recipients with the use of cyclosporine and regimens without cyclosporine that also include azathioprine.12,52-55 However, cyclosporine may have other benefits in comparison with azathioprine. It appears that it is the level
Journal of the American Academy of Dermatology January 1999
of immunosuppression rather than the particular agent used that is important. It is a policy at our center to attempt to reduce or eliminate corticosteroids in those situations with fewer than 3 rejection episodes in the first 3 months after transplantation. Our study confirms the high incidence of skin cancer in the heart transplant population in Australia with a predominance of SCC. SCC and melanoma are a major cause of morbidity and mortality in the long term. We confirm that recipient HLA-DR homozygosity may be a risk factor for skin cancer, whereas HLA-DR7, HLA-A1, and HLA-A11 may be protective factors. Our study is the first to report significant associations between skin cancer occurrence and HLA antigens among heart transplant patients, significant associations having only been previously reported for the kidney transplant population. Our findings reinforce the need to conduct regular skin examinations of heart transplant patients so that lesions can be treated early, given the risk of progressive tumors and metastasis. Patients must be educated regarding the need for sun protection and to regularly carry out self examination. The possible role of retinoids in preventing skin cancer in heart transplant recipients is worthwhile to examine, based on the promising results in kidney transplant recipients. We thank Dr Leonard Ong for assistance and Dr Deborah Black and Associate Professor Judy Simpson for advice regarding statistical analysis. REFERENCES 1. Penn I. Malignancy. Surg Clin North Am 1994;74:124757. 2. Sheil AGR. Malignancy in organ transplant recipients. Transplant Proc 1996;28:1162. 3. Kwa RE, Campana K, Moy RL. Biology of cutaneous squamous cell carcinoma. J Am Acad Dermatol 1992;26: 1-26. 4. Gupta AK, Cardella CJ, Haberman HF. Cutaneous malignant neoplasms in patients with renal transplants. Arch Dermatol 1986;122:1288-93. 5. Koranda FC, Dehmel EM, Kahn G, Penn I. Cutaneous complications in immunosuppressed renal homograft recipients. JAMA 1974;229:419-24. 6. Euvrard S, Kanitakis J, Poutell-Noble C, Disant F, Dureau G, Finaz de Villaine J, et al. Aggressive squamous cell carcinomas in organ transplant recipients. Transplant Proc 1995;27:1767-8. 7. Euvrard S, Kanitakis J, Pouteil-Nobel C, Dureau G, Touraine JL, Faure M, et al. Comparative epidemiologic study of premalignant and malignant epithelial cutaneous lesions developing after kidney and heart transplantation. J Am Acad Dermatol 1995;33:222-9. 8. Sheil AGR, Disney APS, Mathew TH, Livingston BER,
Journal of the American Academy of Dermatology Volume 40, Number 1
9.
10. 11. 12.
13. 14.
15.
16. 17. 18.
19. 20.
21. 22.
23. 24. 25.
26. 27.
Keogh AM. Lymphoma incidence, cyclosporine, and the evolution and major impact of malignancy following organ transplantation. Transplant Proc 1997;29:825-7. Green A, Battistutta D, Hart V, Leslie D, Weedon D, the Nambour Study Group. Skin cancer in a subtropical Australian population: incidence and lack of association with occupation. Am J Epidemiol 1996;144:1034-40. Marks R, Staples M, Giles GG. Trends in nonmelanocytic skin cancer treated in Australia: the second national survey. Int J Cancer 1993;53:585-90. Jelfs PL, Giles G, Shugg D, Coates M, Durling G, Fitzgerald P, et al. Cutaneous malignant melanoma in Australia, 1989. Med J Aust 1994;161:182-7. Bouwes Bavinck JN, Hardie DR, Green A, Cutmore S, MacNaught A, O’Sullivan B, et al. The risk of skin cancer in renal transplant recipients in Queensland, Australia. Transplantation 1996;61:715-21. Espana A, Redondo P, Fernandez AL, Zabala M, Herreros J, Llorens R, et al. Skin cancer in heart transplant recipients. J Am Acad Dermatol 1995;32:458-65. Jensen P, Clausen OPF, Geiran O, Simonsen S, Relbo A, Hansen S, et al. Cutaneous complications in heart transplant recipients in Norway 1983-1993. Acta Derm Venereol (Stockh) 1995;75:400-3. Mihalov HL, Gattuso P, Abraham K, Holmes EW, Reddy V. Incidence of post-transplant malignancy among 674 solid organ transplant recipients at a single centre. Clin Transplant 1996;10:248-55. Walder BK, Robertson MR, Jeremy D. Skin cancer and immunosuppression. Lancet 1971;2:1282-3. Czarnecki D, Collins N, Meehan C, O’Brien T, Leahy S, Nash CA. Squamous cell carcinoma in southern and northern Australia. Int J Dermatol 1992;31:492-3. Marks R, Jolley D, Dorevitch AP, Selwood TS. The incidence of non-melanocytic skin cancers in an Australian population: results of a five-year prospective study. Med J Aust 1989;150:475-8. Green A, Beardmore G, Hart V, Leslie D, Marks R, Staines D. Skin cancer in a Queensland population. J Am Acad Dermatol 1988;19:1045-52. Qunibi W, Akhtar M, Sheth K, Ginn HE, Al-Furayh O, DeVol EB, et al. Kaposi’s sarcoma: the most common tumor after renal transplantation in Saudi Arabia. Am J Med 1988;84:225-32. Greene MH, Young TI, Clark WH Jr. Malignant melanoma in renal-transplant recipients. Lancet 1981;1: 1196-9. Nixon RL, Dorevitch AP, Marks R. Squamous cell carcinoma of the skin: accuracy of clinical diagnosis and outcome of follow-up in Australia. Med J Aust 1986;144: 235-9. Czarnecki D, Staples M, Mar A, Giles G, Meehan C. Metastases from squamous cell carcinoma of the skin in Southern Australia. Dermatology 1994;189:52-4. Moller R, Reyman F, Hou-Jensen K. Metastases in dermatological patients with squamous cell carcinoma. Arch Dermatol 1979;115:703-5. Granstein RD. Photoimmunology. In: Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austen KF, editors. Dermatology in general medicine. New York: McGrawHill; 1993. p. 1638-50. Aubry F, MacGibbon B. Risk factors of squamous cell carcinoma of the skin: a case-control study in the Montreal region. Cancer 1985;55:907-11. Karagas MR, Stukel TA, Greenburg R, Baron JA, Mott LA, Stern RS, et al. Risk of susequent basal cell carcino-
Ong et al 33
28. 29. 30. 31. 32.
33.
34.
35. 36.
37.
38.
39.
40.
41.
42. 43.
44.
ma and squamous cell carcinoma of the skin among patients with prior skin cancer. JAMA 1992;267:330510. Grodstein F, Speizer FE, Hunter DJ. A prospective study of incident squamous cell carcinoma of the skin in the Nurses Health Study. J Natl Cancer Inst 1995;87:1061-6. Wynder ED, Hoffman D. Experimental tobacco carcinogenesis. Adv Cancer Res 1964;8:249-453. Smith JB, Fenske NA. Cutaneous manifestations and consequences of smoking. J Am Acad Dermatol 1996; 34:717-32. Zeid NA, Muller HK. Tobacco smoke condensate cutaneous carcinogenesis: changes in Langerhans’ cells and tumour regression. Int J Exp Pathol 1995;76:75-83. Holt PG, Keast D. Environmentally induced changes in immunological function: acute and chronic effects of inhalation of tobacco smoke and other atmospheric contaminants in man and experimental animals. Bacteriol Rev 1977;41:205-16. Ferson M, Edwards A, Lind A, Milton GW, Hersey P. Low natural killer-cell activity and immunoglobulin levels associated with smoking in human subjects. Int J Cancer 1979;23:603-9. King GN, Healy CM, Glover MT, Kwan JT, Williams DM, Leigh IM, et al. Increased prevalence of dysplastic and malignant lip lesions in renal-transplant recipients. N Engl J Med 1995;332:1052-7. Czarnecki D, Lewis A, Nicholson I, Tait B. Multiple basal cell carcinomas and HLA frequencies in southern Australia. J Am Acad Dermatol 1991;24:559-61. Czarnecki D, Lewis A, Nicholson I, Tait B, Nash C. HLA-DR1 is not a sign of poor prognosis for the development of multiple basal cell carcinomas. J Am Acad Dermatol 1992;26:717-9. Glover MT, Bodmer J, Bodmer W, Kennedy LJ, Brown J, Navarrete C, et al. HLA antigen frequencies in renal transplant recipients and non-immunosuppressed patients with non-melanoma skin cancer. Eur J Cancer 1993;29A:520-4. Bouwes Bavinck JN, Kootte AMM, Van der Woude FJ, Vandenbroucke JP, Vermeer BJ, Claas FHJ. On a possible protective effect of HLA-A11 against skin cancer and keratotic skin lesions in renal transplant recipients. J Invest Dermatol 1991;97:269-72. Czarnecki D, Watkins F, Leahy S, Dyall-Smith D, Lewis A, Nicholson I, et al. Skin cancers and HLA frequencies in renal transplant recipients. Dermatology 1992;185: 9-11. Bouwes Bavinck JN, Claas FHJ, Hardie DR, Green A, Vermeer BJ, Hardie IR. Relation between HLA antigens and skin cancer in renal transplant recipients in Queensland, Australia. J Invest Dermatol 1997;108: 708-11. Bouwes Bavinck JN, Vermeer BJ, van der Woude FJ, Vandenbroucke JP, Geziena M, Schreuder GMT, et al. Relation between skin cancer and HLA antigens in renal transplant recipients. N Engl J Med 1991:325:843-8. Czarnecki DB, Lewis A, Nicholson I, Tait B. Multiple nonmelanoma skin cancer associated with HLA DR7 in Southern Australia. Cancer 1991;68:439-40. Schroeder JS, Gao SZ, Alderman EL, Hunt SA, Johnstone I, Boothroyd DB, et al. A preliminary study of diltiazem in the prevention of coronary artery disease in heart transplant recipients. N Engl J Med 1993;328:164-70. Wagner K, Neumayer H-H. Influence of the calcium antagonist diltiazem on delayed graft function in cadav-
Journal of the American Academy of Dermatology January 1999
34 Ong et al
45.
46. 47.
48. 49.
50.
eric kidney transplantation: results of a 6 month followup. Transplant Proc 1987;19:1353-7. Wahlberg J, Hanas E, Bergstrom C, Fellstrom B, SkarpOrberg I, Frodin L. Diltiazem treatment with reduced dose of cyclosporine in renal transplant recipients. Transplant Proc 1992;24:311-2. McMillen MA, Lewis T, Jaffe BM, Wait RB. Verapamil inhibition of lymphocyte proliferation and function in vitro. J Surg Res 1985;39:76-80. Kelly GE, Meikle W, Sheil AGR. Effects of immunosuppressive therapy on the induction of skin tumours by ultraviolet irradiation in hairless mice. Transplantation 1987;44:429-34. Taylor AEM, Shuster S. Skin cancer after renal transplantation: the causal role of azathioprine. Acta Derm Venereol (Stockh) 1992;72:115-9. Lennard L, Thomas S, Harrington CI, Maddocks JL. Skin cancer in renal transplant recipients is associated with increased concentrations of 6-thioguanine nucleotide in red blood cells. Br J Dermatol 1985;113: 723-9. Hemmens VJ, Moore DE. Photochemical sensitisation by azathioprine and its metabolites: II. Azathioprine and
51.
52.
53.
54.
55.
nitroimidazole metabolites. Photochem Photobiol 1986; 43:257-62. Servilla KS, Burnham DK, Daynes RA. Ability of cyclosporine to promote the growth of transplanted ultraviolet radiation induced tumours in mice. Transplantation 1987;44:291-5. Shuttleworth D, Marks R, Griffin PJA, Salaman JR. Epidermal dysplasia and cyclosporine therapy in renal transplant patients: a comparison with azathioprine. Br J Dermatol 1989;120:551-4. Bunney MH, Benton EC, Barr BB, Smith JW, Anderton JL, Hunter JA. The prevalence of skin disorders in renal allograft recipients receiving cyclosporin A compared with those receiving azathioprine. Nephrol Dial Transplant 1990;5:379-82. Gruber SA, Gillingham K, Sothern RB, Stephanian E, Matas AJ, Dunn DL. De novo cancer in cyclosporinetreated and non-cyclosporine-treated adult primary renal allograft recipients. Clin Transplant 1994;8:388-95. Theil G, Bock A, Spondlin M, Brunner FP, Mihatsch M, Rufli T, et al. Long-term benefits and risks of Cyclosporin A (Sandimmun): an analysis at 10 years. Transplant Proc 1994;26:2493-8.
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the Journal of the American Academy of Dermatology are available to subscribers (only) for the 1999 issues from the Publisher at a cost of $123.00 for domestic, $151.94 for Canadian, and $142.00 for international for volume 40 (January-June) and volume 41 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, Inc., Periodical Subscription Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318. USA: phone (800) 453-4351; (314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription.
Skin cancer is the most common malignancy affecting organ transplant recipients.1,2 Immunosuppressed patients are more predisposed to skin malignancy, which often occurs at an earlier age1-5 and is often multifocal and more aggressive in nature than in nonimmunosuppressed persons.1,2,4,6 Cutaneous malignancy is a significant cause of morbidity and can also be a major cause of malignancy-related mortality. Apart from druginduced immunosuppression and previous sun exposure, human papillomavirus (HPV) and immunogenetic factors (HLA haplotype and patient-donor mismatching) have been reported to play a role in the development of skin cancer. Heart transplant recipients experience a greater rate of skin cancer and other malignancies than kidney transplant recipients.7,8 This has been attributed to the slightly higher doses of immunoFrom the Cardiopulmonary Transplant Unit, St Vincent’s Hospital,a and Skin and Cancer Foundation Australia.b Accepted for publication Sept 15, 1998. Reprint requests: Associate Professor Steven Kossard, Skin and Cancer Foundation, 277 Bourke St, Darlinghurst, NSW, 2010, Australia. Copyright © 1999 by the American Academy of Dermatology, Inc. 0190-9622/99/$8.00 + 0 16/1/94446
suppression agents used in heart transplantation compared with other solid organ transplants. Australia has one of the highest rates of skin malignancy in the world. The rate of nonmelanoma skin cancer for the period 1985-1992 in the most northern state of the country (Queensland) was noted for basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) to be 2074 and 1579 per 100,000 population per year, respectively. SCC occurred at half the rate of BCC among men and at approximately one third the rate among women.9 In 1990, the rate of nonmelanoma skin cancer in northern Australia was up to 4 times that in southern Australia,10 whereas the incidence rate of melanoma for 1989 in Queensland was twice that in southernmost Australia.11 As expected, the rate of skin cancer in kidney transplant recipients in Australia is also one of the highest because previous sun exposure appears to be a major factor in the development of skin cancers. For kidney transplant recipients in Queensland, the rate was noted to be 45% after 11 years, rising to 70% after 20 years of immunosuppression.12 Other countries with high sun exposure such as Spain also have a high cumulative risk for skin cancer of 43.8% after 7 years after heart 27
Journal of the American Academy of Dermatology January 1999
28 Ong et al Table I. Skin lesions and patients affected Total
SCC BCC KA Melanoma Bowen’s disease Total
Caucasian
Non-Caucasian
Patients affected
No. of lesions
No. of patients
No. of lesions
No. of patients
No. of lesions
113 92 19 7 79 152
849 285 28 7 263 1436*
111 91 19 7 75 148
844 278 28 7 256 1417*
2 1 0 0 4 4
5 7 0 0 7 19
KA, Keratoacanthoma. *Includes 3 atypical fibroxanthomas and one Merkel cell tumor.
Table II. Distribution of lesions Location
No. of lesions (%)
Head and neck Trunk Upper limb Lower limb
929 (65) 207 (14) 183 (13) 117 (8)
transplantation.13 There have been no reports to date on the rates of skin cancer in Australian heart transplant recipients. In this study, we examine the rate of skin cancer in Australian heart transplant recipients and its relation to various factors, including HLA antigens. MATERIAL AND METHODS The heart-lung transplant unit at St Vincent’s Hospital in Sydney, Australia performed 620 heart transplants between 1984 and January 1998. This retrospective study examines outcomes in the first 455 patients undergoing transplantation. Information on skin cancers was gathered from several sources, including patients, medical practitioners, medical records, and hospitals in the compilation of the Transplant Unit database. Only histologically confirmed malignancies were included in the analysis. Descriptive statistics were performed on Microsoft Excel 5.0 and inferential statistics performed on Statview 4.5 (Abacus software). The SAS statistical software package (SAS Institute Inc) was used for univariate and multivariate logistic regression, including the stepwise selection procedure. Forty-five patients who died within 6 months after transplant and 10 patients for whom information was incomplete were excluded from the study. As a result 400 patients were studied who experienced a total of 1436 skin lesions (mainly SCC, BCC, keratoacanthoma, Bowen’s disease, and melanoma). The age of
patients ranged from 6.6 to 67 years at transplantation (median, 47.9 years). Fourteen patients were younger than 16 years of age at transplantation. The median follow-up of the 400 patients was 5.5 years (minimum, 0.5 years; maximum 13 years) taken from transplantation to March 1997 or date of death. Two hundred fifteen patients had a follow-up period of more than 5 years. Of the study population, 347 (87%) were of Caucasian background, 31 (8%) of Mediterranean origin, 16 (4%) of Asian origin, and 6 (1.5%) of Aboriginal descent. Immunosuppressive protocols comprised therapy with azathioprine and cyclosporine with or without prednisolone. Rejection episodes of grade greater than 3a in the first 4 weeks after transplantation were treated with intravenous pulsed methylprednisolone and beyond 4 weeks with an oral prednisolone taper. More resistant rejection was treated variously with OKT-3, antithymocyte globulin, total lymphoid irradiation or more recently with conversion from cyclosporine to FK 506 (tacrolimus) or from azathioprine to mycophenolate mofetil. RESULTS
Lesions developed in 152 of 400 patients (38%). A total of 1436 lesions (Table I) were noted with an SCC/BCC ratio of 3:1; a larger proportion of SCC in male than in female patients was found (3.2:1 vs 1.3:1). The majority of lesions occurred on the head and neck, with the least on the lower limb (Table II). The crude cumulative incidence rates of skin cancer were 8.5%, 21%, 31%, and 43% after 1, 3, 5, and 10 years after transplantation, respectively (Kaplan-Meier estimate). Four male patients had more than 50 skin cancers. Metastases developed in 9 male patients with SCC and 4 with melanoma. Eleven deaths from skin cancer (6 from SCC, 4 from melanoma, 1 from Merkel cell carcinoma) accounted for 27% of
Journal of the American Academy of Dermatology Volume 40, Number 1
Ong et al 29
Table III. Multivariate logistic regression models: Odds ratio (95% confidence interval) Factor
Caucasian origin Age (y) at transplantation* Duration (y) of follow-up* HLA-DR7 HLA-DR homozygosity HLA-A1 HLA-A11 Residence in northern Australia
Cancer overall
BCC
SCC
Bowen’s disease
15.1 (4.8-47.3) 1.6 (1.4-1.9) 5.0 (3.2-7.9) 0.4 (0.2-0.7) 2.0 (1.1-3.8) — — —
20.7 (2.8-154.6) 1.5 (1.3-1.8) 2.7 (1.7-4.2) — — — — 2.3 (1.0-5.4)
16.3 (3.6-73.4) 1.7 (1.4-1.9) 6.0 (3.7-9.8) 0.5 (0.2-0.9) — — — —
4.1 (1.3-13.2) 1.6 (1.3-1.9) 6.5 (3.8-11.3) 0.4 (0.2-0.8) — 0.5 (0.3-0.9) 0.4 (0.1-1.0) —
*For being 5 years older or having had a further 5 years of follow-up.
deaths occurring after the fourth year after transplantation. Smoking was not significantly related to the occurrence of skin or lip cancers. The overall incidence of SCC was 37,949 per 100,000 (849/2237.2) posttransplant person-years; that of BCC, 12,739 per 100,000 (285/2237.2) posttransplant person-years; that of Bowen’s disease, 11,755 per 100,000 (263/2237.2) personyears; and that of melanoma, 312 per 100,000 (7/2237.2) person-years. Overall, BCC occurred earlier than SCC after transplantation (4.3 ± 2.6 vs 6.2 ± 2.5 years [mean ± standard deviation]; P < .0001, 2-sample t test). The number of skin cancers was significantly correlated with both age at transplantation (P < .0001, τ 0.168, Kendall rank correlation corrected for ties) and duration of follow-up (P < .0001, τ 0.246). Both were significantly correlated with the number of cancers at different sites and to subtypes of cancer (all P values < .03 except melanoma). Patients with skin cancer were significantly older at transplantation than those with no skin cancer (50.0 vs 45.8 years, median; P < .0001, Mann-Whitney U test). The median age of patients with skin cancer at diagnosis of their first lesion was 54.0 years. On univariate analysis, Caucasian origin (compared with non-Caucasian) was associated with skin cancer: odds ratio 18.5 (95% confidence interval, 2.5-135.6) for BCC occurrence, 12.0 (2.9-50.1) for SCC, and 3.4 (1.2-9.7) for Bowen’s disease. On univariate logistic regression, skin cancer occurrence was more frequent among male than female patients (odds ratio 1.9; 95% confidence interval 1.1-3.3), with all types being more common among male patients but not significantly so. Male patients had a significantly greater
number of SCC (P = .03, Mann-Whitney U test) and total lesions than female patients (P = .02), but only among Caucasian patients. Male and female patients had similar follow-up periods, but male patients were significantly older at transplantation (P = .002) and experienced fewer treated rejection episodes (P = .01). Residing in northern Australia (Northern Territory and Queensland) just before transplantation was significantly associated with all types of skin cancer on univariate analysis (odds ratios, 2.0-2.7). Stepwise selection was used to determine a multivariate logistic regression model for each type of cancer; only variables meeting the .05 significance level were entered into the model. The odds ratios of having a particular type of cancer or cancer overall are listed in Table III. Although skin cancer was associated with being male and residing in northern Australia, these factors did not remain significant in multivariate analysis except for the latter in BCC. We did not find any other HLA antigens, the occurrence of HLA antigen mismatching, or the number of treated rejection episodes to be significantly associated with skin cancer occurrence. Polytomous regression did not provide additional findings. DISCUSSION
Our study confirms the high incidence of skin cancer in organ transplant recipients, as found in the study on Queensland kidney transplant recipients.12 Our high incidence may be partly due to inclusion of lesions other than BCC and SCC that were not included in that study. However, it is more likely that we may have underestimated the incidence of skin cancer because some lesions, in particular BCCs, would have been treated by non-
Journal of the American Academy of Dermatology January 1999
30 Ong et al Table IV. Studies of skin cancer in heart transplant recipients Authors
No. of patients
Euvrard et al7* (France)
150
Espana et al13 (Spain)
92
Age (y) at transplant
55.7 ± 6.2
Follow-up (y)
Findings
Not stated†
136 skin lesions in 41 patients (27%); SCC/BCC ratio: 1.08:1 26 skin cancers in 14 patients (15.2%); SCC/BCC ratio: 1.5:1 96 lesions total; 12.9% had malignant lesions; SCC/BCC ratio: 1:1.08 20 lesions in 20 patients (6.5%); SCC/BCC ratio: 1:1
49.5 (0.17-69)
3.6 (0.1-9.5)
Jensen et al14 (Norway)
140
47.7 (2-63)
5 (1-10)
Mihalov et al15* (US)
307
45.1 (6-71)
2.8
*Other solid organ transplant recipients studied have not been included in this table. †This study examined 150 patients referred for dermatologic problems over a 5-year period.
surgical means and thus would not have been included in the study. The estimates from our study may be lower than some other studies because we have included patients of nonCaucasian background (13.2%) and a proportion of our patients of Angloceltic and European origin had spent their early years in Northern Europe, which would have tended to reduce their cumulative sun exposure. Skin cancers developed in 2 of 6 Aboriginal patients and in 2 of 16 patients of Mediterranean origin. Logistic regression showed that patients of Caucasian origin and those residing in northern Australia just before transplantation had a higher frequency of skin cancers, although the latter remained significant only for BCC in multivariate analysis. The finding of a 3:1 SCC/BCC ratio is consistent with previous studies (Table IV) in which a reversal in the SCC/BCC ratio was noted when compared with the general population.1,7,13 In kidney transplant recipients, the SCC/BCC ratio has been noted to be up to 16:1.16 The greater SCC/BCC ratio for male patients as compared with female patients (3.2:1 vs 1.3:1) can be explained by the greater number of lesions male patients have after transplantation. On multivariate analysis, although there was a strong trend for some skin cancers, sex did not remain a significant factor, indicating that other factors may be more important. Mihalov et al15 found that SCC occurred later than BCC in heart transplant patients, whereas the opposite was true of kidney transplant patients. In addition, BCC occurred earlier and SCC later in heart than in kidney transplant patients. In contrast, Euvrard et al7 found that skin lesions occurred earlier in heart transplant patients, but
BCC did comprise a higher proportion of lesions in their heart transplant patients. In our study, when all SCCs and BCCs are considered, the interval to the development of SCC was significantly longer than for BCC (6.2 ± 2.5 vs 4.3 ± 2.6 years, mean ± SD; P < .0001, 2-sample t test). It appears that BCCs occur earlier after transplantation and SCCs develop and predominate after a longer period of immunosuppression. The correlation of the number of skin cancers with age at transplantation and the duration of follow-up, as well as the odds ratios reported for their occurrence, is consistent with the hypothesis that exposure to sunlight before transplantation and duration of exposure to immunosuppression are major risk factors. Our study also reinforces the concept that immunosuppressed patients have skin cancers at an earlier age. Patients had a median age of 54 years at diagnosis of their first histologically confirmed skin cancer, whereas in the general population the average age of patients with either SCC or BCC is 10 years older,17-19 although those living closer to the equator develop lesions at a considerably younger age.17 There were no cases of Kaposi’s sarcoma in our study, in contrast to the Australian kidney transplant population2 and transplant recipients overseas for whom rates up to 5.3% in Saudi Arabia have been reported.1,20 This can be explained in part by the ethnicity of our transplant population of which 92% were of non-Mediterranean origin and thus not in a high-risk group for Kaposi’s sarcoma. Merkel cell carcinoma developed in 1 patient, atypical fibroxanthoma in 3 patients, leiomyosarcoma in 1, and in 2 patients, lymphoproliferative disease presented initially as skin lesions.
Journal of the American Academy of Dermatology Volume 40, Number 1
In our group, 11 patients died of skin cancer, accounting for 27% of deaths occurring after the fourth year post transplant. Seven patients had melanomas; metastases developed in 4 of these patients and they died of melanoma. In one study, the majority of melanomas occurring after kidney transplantation arose in precursor nevi,21 indicating the need to monitor nevi closely after transplantation. Metastases developed in 9 patients with SCC; 6 died of metastatic SCC. The patient with Merkel cell carcinoma died of metastatic disease. Invasion of the parotid gland developed in 4 patients with SCC and invasion of the external ear requiring radiotherapy developed in 2 patients, highlighting the role of immunosuppression in predisposing patients to more aggressive cancers. The risk of metastases in SCC is estimated to be 1% to 2% in the general population,22 although a rate of 5.8% was found in a hospital-based study.23 Cutaneous SCC secondary to inflammatory and degenerative processes (ostemyelotic foci, burns, scars of lupus vulgaris) developed metastases in 10% to 30% of such cases. Eleven percent of labial SCC metastasize in contrast to 3.3% of skin SCC.24 Apart from origin, anatomic site, immunosuppression, histologic features (such as tumor grade and perineural invasion), size and depth of the tumor, and previous treatment also influence the risk of metastasis.3 In our study, metastases subsequently developed in 8% of patients with SCC, contributed to a large extent by the occurrence of multiple lesions in patients. SCCs, which are the nonmelanoma skin cancer most likely to metastasize, are also the most poorly diagnosed clinically.22 Nixon, Dorevitch, and Marks22 recommend biopsy as a routine to confirm the clinical diagnosis, particularly if the tumor’s histologic type will influence the mode of its removal. The majority of skin cancers occurring on the head and neck supports the role of sun exposure in causation. UV light radiation is not only a carcinogen but can induce tolerance or immunosuppression to tumor antigens by interfering with Langerhans cell antigen presentation.25 Seventy percent of 152 patients with skin cancer on whom a smoking history was available were ex-smokers, suggesting that smoking may contribute to skin cancer as it does in labial, lung, and cervical cancer. Smoking is a risk factor for SCC,26-28 and tobacco smoke condensate can induce SCC in rabbit and mouse skin.29 Smokers with melanoma are
Ong et al 31 more likely to have metastases on initial presentation and have a poorer prognosis than nonsmokers.30 Tobacco smoke condensate can impair function of Langerhans cells31 and affect other parts of the immune system as well.32,33 Both sun exposure and smoking are risk factors for dysplastic and malignant lip lesions in kidney transplant recipients.34 Although we did not find an association between smoking and skin or lip cancer, our results are not conclusive because we have not quantified the extent of smoking by individual patients. Many studies have examined immunogenetic factors, particularly HLA haplotypes in skin cancer. In normal nonimmunosuppressed patients, HLA-DR4 was found to be significantly less common in patients with multiple BCCs, whereas HLA-DR1 was more common.35-37 In transplant recipients, HLA-A11 was found to be protective against skin cancer38 in The Netherlands. However, this was not found in kidney transplant recipients in Southern Australia39 and was associated with skin cancer in northern Australia.40 These differences are possibly related to the relative importance of human papillomavirus or sun exposure in skin cancer development.40 Other antigens found to be associated with skin cancer in kidney transplant recipients include HLADQw2,37 HLA-DR7,39,40 HLA-A3,38 and HLAB27.38,39 Varied findings of Bouwes Bavinck et al38 and Czarnecki et al35,36,39 may be due to the effect of geographic location on the incidence of skin cancer in the transplant and general population. However, both found that HLA-B2738,39 was a strong risk factor for skin cancer in the kidney transplant population. In addition, Bouwes Bavinck et al41 have found that HLA-B mismatching and HLA-DR homozygosity are significantly associated with an increased risk of SCC. In our study, multivariate logistic regression models arising from stepwise selection show that the occurrence of skin cancer is associated with HLA-DR homozygosity. This is consistent with a study of kidney transplant recipients in The Netherlands41 but was not found in northern Australia.40 Our findings that absence of HLADR7 is associated with the occurrence of skin cancer overall, SCC, and Bowen’s disease is in contrast to findings of previous studies.39,40 However, in one of those studies the finding did not reach significance40 and in the other39 the control group
32 Ong et al was a healthy population, rather than kidney transplant recipients. It should be noted that HLA-DR7 is associated with having multiple BCCs in the nontransplant population.42 We did not find a significant relation between BCC occurrence and presence or absence of HLA-DR7. Our finding that HLA-11 is a protective factor in Bowen’s disease is consistent with that of other studies.38 HLA-A1 as a protective factor for Bowen’s disease has not been previously reported. Although our findings are consistent with HLA antigens affecting the immune response to tumor and/or papillomavirus antigens, it is clear that the association of skin cancer and particular HLA antigens is complex because multiple other factors play a role in skin cancer development. We found no relation between the occurrence of skin cancer and the number of treated rejection episodes on univariate and multivariate analysis. We have not investigated the relation of skin cancer to individual immunosuppressive drugs. Any such analysis should consider drug interactions in assessing the level of immunosuppression. Ketoconazole is often used to potentiate the effect of cyclosporine. Diltiazem, which is used to prevent transplant coronary artery disease,43 affects cyclosporine blood levels,44 reducing dosage requirements by one third45 and may inhibit the proliferation of T lymphocytes.46 Early studies suggested azathioprine had a particular risk of inducing keratotic lesions and skin cancer47,48 possibly related to its active metabolite 6-thioguanine, which is a tumor initiator and promotor.48 Lennard et al49 found the level of this metabolite to be associated with the presence of dysplastic lesions. Alternatively, it was thought that metabolites of azathioprine, the imidazole compounds, may sensitize the skin to sunlight, thereby increasing the risk of skin cancer.50 It was then proposed that cyclosporine should be used instead of azathioprine given the perceived risk with azathioprine. On the other hand, cyclosporine may promote the progression of UV radiation–induced skin cancers.51 However, it has been shown that there is no significant difference in skin cancer incidence in kidney transplant recipients with the use of cyclosporine and regimens without cyclosporine that also include azathioprine.12,52-55 However, cyclosporine may have other benefits in comparison with azathioprine. It appears that it is the level
Journal of the American Academy of Dermatology January 1999
of immunosuppression rather than the particular agent used that is important. It is a policy at our center to attempt to reduce or eliminate corticosteroids in those situations with fewer than 3 rejection episodes in the first 3 months after transplantation. Our study confirms the high incidence of skin cancer in the heart transplant population in Australia with a predominance of SCC. SCC and melanoma are a major cause of morbidity and mortality in the long term. We confirm that recipient HLA-DR homozygosity may be a risk factor for skin cancer, whereas HLA-DR7, HLA-A1, and HLA-A11 may be protective factors. Our study is the first to report significant associations between skin cancer occurrence and HLA antigens among heart transplant patients, significant associations having only been previously reported for the kidney transplant population. Our findings reinforce the need to conduct regular skin examinations of heart transplant patients so that lesions can be treated early, given the risk of progressive tumors and metastasis. Patients must be educated regarding the need for sun protection and to regularly carry out self examination. The possible role of retinoids in preventing skin cancer in heart transplant recipients is worthwhile to examine, based on the promising results in kidney transplant recipients. We thank Dr Leonard Ong for assistance and Dr Deborah Black and Associate Professor Judy Simpson for advice regarding statistical analysis. REFERENCES 1. Penn I. Malignancy. Surg Clin North Am 1994;74:124757. 2. Sheil AGR. Malignancy in organ transplant recipients. Transplant Proc 1996;28:1162. 3. Kwa RE, Campana K, Moy RL. Biology of cutaneous squamous cell carcinoma. J Am Acad Dermatol 1992;26: 1-26. 4. Gupta AK, Cardella CJ, Haberman HF. Cutaneous malignant neoplasms in patients with renal transplants. Arch Dermatol 1986;122:1288-93. 5. Koranda FC, Dehmel EM, Kahn G, Penn I. Cutaneous complications in immunosuppressed renal homograft recipients. JAMA 1974;229:419-24. 6. Euvrard S, Kanitakis J, Poutell-Noble C, Disant F, Dureau G, Finaz de Villaine J, et al. Aggressive squamous cell carcinomas in organ transplant recipients. Transplant Proc 1995;27:1767-8. 7. Euvrard S, Kanitakis J, Pouteil-Nobel C, Dureau G, Touraine JL, Faure M, et al. Comparative epidemiologic study of premalignant and malignant epithelial cutaneous lesions developing after kidney and heart transplantation. J Am Acad Dermatol 1995;33:222-9. 8. Sheil AGR, Disney APS, Mathew TH, Livingston BER,
Journal of the American Academy of Dermatology Volume 40, Number 1
9.
10. 11. 12.
13. 14.
15.
16. 17. 18.
19. 20.
21. 22.
23. 24. 25.
26. 27.
Keogh AM. Lymphoma incidence, cyclosporine, and the evolution and major impact of malignancy following organ transplantation. Transplant Proc 1997;29:825-7. Green A, Battistutta D, Hart V, Leslie D, Weedon D, the Nambour Study Group. Skin cancer in a subtropical Australian population: incidence and lack of association with occupation. Am J Epidemiol 1996;144:1034-40. Marks R, Staples M, Giles GG. Trends in nonmelanocytic skin cancer treated in Australia: the second national survey. Int J Cancer 1993;53:585-90. Jelfs PL, Giles G, Shugg D, Coates M, Durling G, Fitzgerald P, et al. Cutaneous malignant melanoma in Australia, 1989. Med J Aust 1994;161:182-7. Bouwes Bavinck JN, Hardie DR, Green A, Cutmore S, MacNaught A, O’Sullivan B, et al. The risk of skin cancer in renal transplant recipients in Queensland, Australia. Transplantation 1996;61:715-21. Espana A, Redondo P, Fernandez AL, Zabala M, Herreros J, Llorens R, et al. Skin cancer in heart transplant recipients. J Am Acad Dermatol 1995;32:458-65. Jensen P, Clausen OPF, Geiran O, Simonsen S, Relbo A, Hansen S, et al. Cutaneous complications in heart transplant recipients in Norway 1983-1993. Acta Derm Venereol (Stockh) 1995;75:400-3. Mihalov HL, Gattuso P, Abraham K, Holmes EW, Reddy V. Incidence of post-transplant malignancy among 674 solid organ transplant recipients at a single centre. Clin Transplant 1996;10:248-55. Walder BK, Robertson MR, Jeremy D. Skin cancer and immunosuppression. Lancet 1971;2:1282-3. Czarnecki D, Collins N, Meehan C, O’Brien T, Leahy S, Nash CA. Squamous cell carcinoma in southern and northern Australia. Int J Dermatol 1992;31:492-3. Marks R, Jolley D, Dorevitch AP, Selwood TS. The incidence of non-melanocytic skin cancers in an Australian population: results of a five-year prospective study. Med J Aust 1989;150:475-8. Green A, Beardmore G, Hart V, Leslie D, Marks R, Staines D. Skin cancer in a Queensland population. J Am Acad Dermatol 1988;19:1045-52. Qunibi W, Akhtar M, Sheth K, Ginn HE, Al-Furayh O, DeVol EB, et al. Kaposi’s sarcoma: the most common tumor after renal transplantation in Saudi Arabia. Am J Med 1988;84:225-32. Greene MH, Young TI, Clark WH Jr. Malignant melanoma in renal-transplant recipients. Lancet 1981;1: 1196-9. Nixon RL, Dorevitch AP, Marks R. Squamous cell carcinoma of the skin: accuracy of clinical diagnosis and outcome of follow-up in Australia. Med J Aust 1986;144: 235-9. Czarnecki D, Staples M, Mar A, Giles G, Meehan C. Metastases from squamous cell carcinoma of the skin in Southern Australia. Dermatology 1994;189:52-4. Moller R, Reyman F, Hou-Jensen K. Metastases in dermatological patients with squamous cell carcinoma. Arch Dermatol 1979;115:703-5. Granstein RD. Photoimmunology. In: Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austen KF, editors. Dermatology in general medicine. New York: McGrawHill; 1993. p. 1638-50. Aubry F, MacGibbon B. Risk factors of squamous cell carcinoma of the skin: a case-control study in the Montreal region. Cancer 1985;55:907-11. Karagas MR, Stukel TA, Greenburg R, Baron JA, Mott LA, Stern RS, et al. Risk of susequent basal cell carcino-
Ong et al 33
28. 29. 30. 31. 32.
33.
34.
35. 36.
37.
38.
39.
40.
41.
42. 43.
44.
ma and squamous cell carcinoma of the skin among patients with prior skin cancer. JAMA 1992;267:330510. Grodstein F, Speizer FE, Hunter DJ. A prospective study of incident squamous cell carcinoma of the skin in the Nurses Health Study. J Natl Cancer Inst 1995;87:1061-6. Wynder ED, Hoffman D. Experimental tobacco carcinogenesis. Adv Cancer Res 1964;8:249-453. Smith JB, Fenske NA. Cutaneous manifestations and consequences of smoking. J Am Acad Dermatol 1996; 34:717-32. Zeid NA, Muller HK. Tobacco smoke condensate cutaneous carcinogenesis: changes in Langerhans’ cells and tumour regression. Int J Exp Pathol 1995;76:75-83. Holt PG, Keast D. Environmentally induced changes in immunological function: acute and chronic effects of inhalation of tobacco smoke and other atmospheric contaminants in man and experimental animals. Bacteriol Rev 1977;41:205-16. Ferson M, Edwards A, Lind A, Milton GW, Hersey P. Low natural killer-cell activity and immunoglobulin levels associated with smoking in human subjects. Int J Cancer 1979;23:603-9. King GN, Healy CM, Glover MT, Kwan JT, Williams DM, Leigh IM, et al. Increased prevalence of dysplastic and malignant lip lesions in renal-transplant recipients. N Engl J Med 1995;332:1052-7. Czarnecki D, Lewis A, Nicholson I, Tait B. Multiple basal cell carcinomas and HLA frequencies in southern Australia. J Am Acad Dermatol 1991;24:559-61. Czarnecki D, Lewis A, Nicholson I, Tait B, Nash C. HLA-DR1 is not a sign of poor prognosis for the development of multiple basal cell carcinomas. J Am Acad Dermatol 1992;26:717-9. Glover MT, Bodmer J, Bodmer W, Kennedy LJ, Brown J, Navarrete C, et al. HLA antigen frequencies in renal transplant recipients and non-immunosuppressed patients with non-melanoma skin cancer. Eur J Cancer 1993;29A:520-4. Bouwes Bavinck JN, Kootte AMM, Van der Woude FJ, Vandenbroucke JP, Vermeer BJ, Claas FHJ. On a possible protective effect of HLA-A11 against skin cancer and keratotic skin lesions in renal transplant recipients. J Invest Dermatol 1991;97:269-72. Czarnecki D, Watkins F, Leahy S, Dyall-Smith D, Lewis A, Nicholson I, et al. Skin cancers and HLA frequencies in renal transplant recipients. Dermatology 1992;185: 9-11. Bouwes Bavinck JN, Claas FHJ, Hardie DR, Green A, Vermeer BJ, Hardie IR. Relation between HLA antigens and skin cancer in renal transplant recipients in Queensland, Australia. J Invest Dermatol 1997;108: 708-11. Bouwes Bavinck JN, Vermeer BJ, van der Woude FJ, Vandenbroucke JP, Geziena M, Schreuder GMT, et al. Relation between skin cancer and HLA antigens in renal transplant recipients. N Engl J Med 1991:325:843-8. Czarnecki DB, Lewis A, Nicholson I, Tait B. Multiple nonmelanoma skin cancer associated with HLA DR7 in Southern Australia. Cancer 1991;68:439-40. Schroeder JS, Gao SZ, Alderman EL, Hunt SA, Johnstone I, Boothroyd DB, et al. A preliminary study of diltiazem in the prevention of coronary artery disease in heart transplant recipients. N Engl J Med 1993;328:164-70. Wagner K, Neumayer H-H. Influence of the calcium antagonist diltiazem on delayed graft function in cadav-
Journal of the American Academy of Dermatology January 1999
34 Ong et al
45.
46. 47.
48. 49.
50.
eric kidney transplantation: results of a 6 month followup. Transplant Proc 1987;19:1353-7. Wahlberg J, Hanas E, Bergstrom C, Fellstrom B, SkarpOrberg I, Frodin L. Diltiazem treatment with reduced dose of cyclosporine in renal transplant recipients. Transplant Proc 1992;24:311-2. McMillen MA, Lewis T, Jaffe BM, Wait RB. Verapamil inhibition of lymphocyte proliferation and function in vitro. J Surg Res 1985;39:76-80. Kelly GE, Meikle W, Sheil AGR. Effects of immunosuppressive therapy on the induction of skin tumours by ultraviolet irradiation in hairless mice. Transplantation 1987;44:429-34. Taylor AEM, Shuster S. Skin cancer after renal transplantation: the causal role of azathioprine. Acta Derm Venereol (Stockh) 1992;72:115-9. Lennard L, Thomas S, Harrington CI, Maddocks JL. Skin cancer in renal transplant recipients is associated with increased concentrations of 6-thioguanine nucleotide in red blood cells. Br J Dermatol 1985;113: 723-9. Hemmens VJ, Moore DE. Photochemical sensitisation by azathioprine and its metabolites: II. Azathioprine and
51.
52.
53.
54.
55.
nitroimidazole metabolites. Photochem Photobiol 1986; 43:257-62. Servilla KS, Burnham DK, Daynes RA. Ability of cyclosporine to promote the growth of transplanted ultraviolet radiation induced tumours in mice. Transplantation 1987;44:291-5. Shuttleworth D, Marks R, Griffin PJA, Salaman JR. Epidermal dysplasia and cyclosporine therapy in renal transplant patients: a comparison with azathioprine. Br J Dermatol 1989;120:551-4. Bunney MH, Benton EC, Barr BB, Smith JW, Anderton JL, Hunter JA. The prevalence of skin disorders in renal allograft recipients receiving cyclosporin A compared with those receiving azathioprine. Nephrol Dial Transplant 1990;5:379-82. Gruber SA, Gillingham K, Sothern RB, Stephanian E, Matas AJ, Dunn DL. De novo cancer in cyclosporinetreated and non-cyclosporine-treated adult primary renal allograft recipients. Clin Transplant 1994;8:388-95. Theil G, Bock A, Spondlin M, Brunner FP, Mihatsch M, Rufli T, et al. Long-term benefits and risks of Cyclosporin A (Sandimmun): an analysis at 10 years. Transplant Proc 1994;26:2493-8.
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the Journal of the American Academy of Dermatology are available to subscribers (only) for the 1999 issues from the Publisher at a cost of $123.00 for domestic, $151.94 for Canadian, and $142.00 for international for volume 40 (January-June) and volume 41 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, Inc., Periodical Subscription Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318. USA: phone (800) 453-4351; (314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription.