- Email: [email protected]
Skin cancer in heart transplant recipients
Skin cancer in heart transplant recipients Agustin Espafia, MD,a Pedro Reclondo, MD,a Angel L. Fernandez, MD,b Martin Zabala, MD,b Jestis Herreros, MD,b Rafael Llorens, MD,b and Emilio Quintanilla, MDa Pumplonu, Spain Background: The frequency of skin cancer in organ transplant recipients is high, up to 15%. Objective: Our purpose was to determine the incidence of skin cancer in patients who underwent immunosuppression after heart transplantation and to determine the factors important in the appearance of skin cancer. Methods: We studied the frequency of skin cancer in 92 of 111 patients after they underwent heart transplantation between January 1984 and December 1993. Results: At least one cutaneous neoplasm (squamous cell carcinoma and/or basal cell carcinoma) developed in 14 patients (15.2%). The basal cell carcinoma to squamous cell carcinoma ratio was 1:1.5. The skin cancer appeared an average of 3 1.5 months after transplantation; the average was 36 months for squamous cell carcinoma and 25.3 months for basal cell carcinoma. Cumulative risk rose from 4.3% at 1 year up to 43.8% at 7 years after transplantation. The overall incidence of both types of skin cancer was 45.3 per 1000 posttransplant person-years, with an incidence of 25.8 for basal cell carcinoma and 29.1 for squamous cell carcinoma. Most skin cancers developed between 2 and 3 years after transplantation. All patients were exposed to a significant amount of ultraviolet radiation and had skin type II or III. We did not find a significant association between skin cancer and haplotype HLA-A3, HLA-Al 1, HLA-DR, and the number of mismatches for HLA-B. Conclusion: We found an increased progressive cumulative incidence of skin cancer in heart transplant recipients for two reasons: (1) immunosuppression and increased exposure to ultraviolet radiation in some patients, and (2) the skin type of certain patients. We emphasize the need for photoprotection in this group of patients and regular skin cancer screening examinations. (J AM ACAD DERMATOL 1995;32:458-65.) Skin manifestations in patients who have undergone organ transplantation are frequent.’ The literature is sparse on cutaneous findings,2-5 especially skin cancer, in patients who have had a heart transplant.6-9 We determined the occurrence of skin cancer in heart transplant patients and compared our results with those in the literature. MATERIAL
AND METHODS
We determined the presence of skin cancer (basal cell carcinoma and squamous cell carcinoma) in 92 of 111 patients who had undergone heart transplantation from January 1984 to December 1993. Like others authors,7 From the Departments of Dermatologya University Clinic of Navarra, School Accepted
for publication
and Cardiovascular of Medicine.
Surgery,b
Sept. 12, 1994
Reprint requests: Agustin Espafia, MD, Department of Dermatology, University Clinic of Navarra, School of Medicine, P.O. Box 192, 31080 Pamplona, Spain. Copyright
@ 1994 by the American
0190-9622/95/$3.00+0
458
16/l/60703
Academy
of Dermatology,
Inc.
we did not include 16 patients who died within 6 months after undergoing transplantation and three patients with a follow-up period of lessthan 3 months after undergoing transplantation. All patients came from the northern and eastern regions of Spain (latitude 40” to 45’ N). Patients ranged in age from 2 months to 69 years (average, 49.5 years). Seven patients were lessthan 15 years of age. The average follow-up period of the 111 patients was 43.3 months (range, 1 to 114 months). Twenty-two patients died. Data from 1984 to 1990 were obtained from medical records. Two dermatologists examined the patients every 6 months, concentrating particularly on the areas of their bodies exposed to sunlight. All suspicious lesions were biopsied. Simple excision was used to treat all patients with a basal cell carcinoma or squamous cell carcinoma. In one patient with squamous cell carcinoma of the lower lip, cervical lymphadenectomy and radiotherapy were also performed. The patient died of generalized metastases. Photoprotective measures were initiated in all patients. The skin type I to VI was assessedaccording to the standard classification9 Details of exposure to ultraviolet radiation in each patient were obtained from the medical
Journal of the American Academy of Dermatology Volume 32, Number 3
Table I. Epidemiologic skin
cancer
459
data
Patient No.
Time of appearance (mo since- transplant)
%e m
BCC BCC BCC BCC BCC
69 38 38 68 55
see see see see see see
6 7 8 9 10 11
64 61 53 53 67 62
BCC + SCC
12
57
BCC + SCC
13
57
BCC + SCC
14
50
BCC, Basal cell carcinoma; *Basal cell carcinoma. fSquamous cell carcinoma.
Espaiia et al.
SCC, squamous
7 36 11 20 10 21 26 26 13 6 72 53 22 24 24 36” 36” 367 26* 41* 26t 31”1‘ 60* 50t 58 547
Skill Warts
-
+ + +
-
type
II III III III II
-
II II II II II II
+
II
-
II
+
II
-
LOCAiOll
Face Face Back Face Face Ear Forehead Scalp Arm Lower lip Lower lip Ear lobe Lower lip Forehead Forehead Face Back Face Face Face Face Face Scalp Lower lip Ear lobe Neck
Profession
Farmer Civil servant Civil servant Businessman Businessman
Civil servant Driver Civil servant Distributor Farmer Civil servant Businessman Bricklayer
Farmer
cell carcinoma
records, following the criteria established by McLelland et al.‘O HLA-A and HLA-B typing was performed in all patients before transplantation, and HLA-DR was determined in 95 patients. t 1,t2 The degree of mismatching was assessedby counting the antigens present in the donor but absent in the recipient. We performed correlations only with HLA-A3, HLA-Al 1, HLA-B27, HLA-DR homozygosity, and HLA-B mismatching, because these are the only haplotypes that have been related to the occurrence of neoplasms in transplant recipients.13-t5 The incidence of skin cancer development (basal cell carcinoma and squamous cell carcinoma) after transplantation per 1000 person-years was computed16 for the overall follow-up period and for the periods of 0 to 12,13 to 24, 25 to 36, 37 to 48, 49 to 60, 61 to 72, and longer than 72 months after transplantation. We also computed the cumulative incidence of skin cancer by life-table analysis.l7 For statistical purposes, data for a patient were included in a given period if he or she had been observed for that entire period or at least for two thirds of that period. When this was not the case, the data for this patient
were included in the previous period, if this situation OG curred during the first year, the data were not taken into account. Fisher’s test was used to assessfor statistically significant relationships between HLA matching and the presence of skin cancer. The following immunosuppressive protocols were used for all patients in each period: 1984 to 1986: Cyclosporine (6 mg/kg before surgery and, according to blood levels, between 200 and 300 rig/ml after surgery); 6-methylprednisolone (500 mg intravenously intraoperatively, and 125 mg three times a day intravenously for 3 days); and oral prednisone (1 mg/kg/day after surgery, reduced to 0.2 mg/kg/day at the end of the fourth week). In patients with renal insufficiency after surgery (creatinine > 1.5), cyclosporine was replaced by equine antithymocytic y-globulin (10 to 20 mg/kg/day) for 7 days. 1986 to 1989: Azathioprine (2 mg/kg intravenously before surgery and 1.5 to 2 mg/kg/day after surgery) and corticosteroid administration as used in patients from 1984 to 1986. Cyclosporine was administered on the third day after surgery and afterward according to blood lev-
460
Journal of the American Academy of Dermatology March 1995
Espaiia et al. 50
Table II. Age/skin cancerdistribution
1
Skin cancer Age w*
BCC
20-29 30-39 40-49 50-59 60-70 Total
1 1 2 5
SCC
-
-
2 4 is
3 5
KC, Basal cell carcinoma; SCC, squamous *Average age of all patients = 49.5 years. 0
2
YEARS
AFTER
4
6
BCC + SCC
cell carcinoma.
a
TRANSPLANTATION
Fig. 1. Cumulative
incidence of total skin cancer, squamous cellcarcinoma,and basal cell carcinoma. HoZlow square, Basal cell carcinoma; solid square, total skin cancer;diamond, squamouscellcarcinoma.
Table III. Relation betweenfollow-up time and type of skin cancer Posttransplant period No. of No. of patients No. of patients with see (mo) I patients I with BCC
o-12 13-24 25-36 37-48 49-60 61-72 >72
els. The same method as that used from 1984 to 1986 was followed with respect to patients with renal insufficiency. I989 to 1993: The protocol was the same as that used for patientsfrom 1986to 1989;however, in patientswith
creatininegreaterthan1S, cyclosporine wassubstituted for OKT3 (5 mg/dayintravenously for 12days).
KC,
Basal cell carcinoma;
92 85 65 51 34 26 16 SCC, squamous
3 2 6 1 1 0 0
1 4 3 0 ‘4 1 0 cell carcinoma.
RESULTS
In 14 patients (15.2%), 26 skin cancerswere found. Five patients had basal cell carcinoma, six had squamouscell carcinoma,and three had both basalcell carcinomaand squamouscell carcinoma (Table I). The basalcell carcinomato squamouscell carcinomaratio was 1:1.5 for the first malignancy. When all secondand subsequentskin cancerswere included,the basalcell carcinomato squamouscell carcinoma ratio was 1:1.3. In four of the patients with squamouscell carcinoma,the lesiondeveloped on the lower lip (at the vermilion border exclusively). All patientshad skin type II or III (Table I). In the first phaseof data gatheringwe identifiedthree patients with skin cancer (patients 2, 6, and 12), whereas11 patientswith skin cancerwere detected in the prospectivefollow-up (patients 1,3,4,5,7,8, 9, 10, 11, 13,and 14) (Table I). Table II showsthe agedistribution of the 14 patientsin whom one or more skin neoplasmsoccurred. Twelve patients (11.5%) had common warts, which appearedbetween 2 and 55 months after transplantation(mean,21.7months).All wartswere
locatedin areasexposedto ultraviolet radiation,and they respondedto cryotherapy.Five of these12 patients had an associatedneoplasm or neoplasms (patient 2, basalcell carcinoma;patient4, basalcell carcinoma;patient 5, basalcell carcinoma;patient 12, basal cell carcinoma and squamouscell carcinoma; and patient 14, basal cell carcinoma and squamouscell carcinoma) (Table I). Overall, skin cancerappearedan averageof 31.5 months after transplantation;basalcell carcinoma appearedat an averageof 25.3 months, and the squamouscell carcinoma appearedat 36 months. Most cancersappearedin the first 4 years after transplantation(Table III). The overallincidenceof both typesof skin cancer was45.3casesper 1000posttransplantperson-years: the incidencewas25.88for basalcell carcinomaand 29.12for squamouscell carcinoma.The overallrisk of developingskin cancerincreasedfrom a cumulative incidenceof 4.3% 1year after transplantationto 43.8% 7 years after transplantation.For basal cell carcinomathe incidenceswere 3.26% after 1 year
Journal of the American Academy of Dermatology Volume 32, Number 3
Espahaet al. 461
Table IV. HLA distribution Type of skin cancer HLA
study
No. of patients with skin cancer
SCC
1 4
4 2
2
19 72
0 4
0 7
1
2
6 85
0 4
0 7
0 3
90
0
2
1
4
5
2
0
0
3
4 3
0 2 P
A3 (NS) Positive 6 Negative 8 All (NS) Positive 1 Negative 13 B27 (NS) Positive 0 Negative 14 DR homozygosity (NS) Positive 3 Negative 11 No. mismatches for HLA-B (NS) 0 0 1 9
2 KC,
Basal cell carcinoma;
5 SCC, squamous
1 cell carcinoma;
BCC + SCC
No. patients without skin cancer
BCC
1
1
38 45 1 18
72
NS, not significant.
and 16.24%after 7 years, and for squamouscell carcinomathe incidenceswere 1.08%after 1 year and 27.55%after 7 years (Fig. 1). The HLA distributions are shownin Table IV. We did not find a significant associationbetween skincancerandtheseHLAs. We did not observeany casesof Kaposi’ssarcomaor malignant melanoma. DISCUSSION Skin canceraccountsfor mostof the neoplasmsin transplantrecipients.2~9l1*, 19Overall,the frequency of skin cancerin patientstaking conventionalimmunosuppressiveagentsis 40%; the frequency is only 15%whencyclosporineis administered.l8 Nevertheless,Blohme and Larko2’ found no differences in the frequencyof skin cancerin patientsreceiving azathioprine,cyclosporine,or combinedtherapy.In other studies with patients who did not receive treatment with cyclosporine,the frequencyof skin cancerrangedfrom 0.5%to 15.6%.21> 22In patients who receivedtreatment with cyclosporine,the frequencyrangedbetween0% to 11%.21 9 Our results, which indicate a frequencyof 15.2%,are in agreement with thesefindings. In our area of Spain the frequencyof skin neoplasmsin the population at large is 1.8%F3approximately one seventhof the figure obtainedin this series. Little information is available about cutaneous neoplasmsin patientsnot receivingtransplantswho
were treated with cyclosporinefor other diseases. Krupp and Monka observeda frequencyof 0.7%, with a basalcell carcinomato squamouscell carcinoma ratio of 1:6.However,all of their patientshad previouslybeentreatedwith PUVA andmethotrexate. These authors believedthat the cancersthey observedwereunrelatedto cyclosporine,becausethe patientshad beentaking this drug for only a short time. No referencehasbeenmadeto skin cancerin the context of other autoirnmune diseases.25 In the statistical analysis of our series,an increasedcumulative incidencewas noted (Fig. 1). Our frequencyis slightly greaterthan that reported by Hardie et a1.,2who found a cumulative risk of 38%after 8 years.In our seriesthe degreeof risk was 43.8% after 7 years. The reason for this high
incidencecouldbethat both studieswerecarriedout in countrieswith high levelsof ultraviolet radiation. In othercountrieswith lesssolarradiation,this risk doesnot reach 10% after 7 years.26Similarly, we found an incidenceof 45.3 casesper 1000personyears,which resemblesthat reportedin othercountries with high levels of ultraviolet radiation (43 cases per 1000 person-years)and exceedsthat encounteredin countries with less sunlight (nine casesper 1000person-years). The basalcell carcinomato squamouscell carcinoma ratio in Latin countrieshasbeenreportedas 2.1 . .23 Among residents of sunny countries who
462
Journal of the American Academy of Dermatology March 1995
Espaiiu et al.
Table V. Age/skin cancerdistribution in patientswith classicimmunosuppressivetreatment Hardie w
20-30 31-40 41-50 51-60 >61 Total
BCC
see
-
-
1 3 _ 4
et al.**
4 7 -
_ 11
BCC, Basal cell carcinoma; XC, squamous *Average age of all patients = 30 years. TAverage age of all patients = 36 years. SAverage age of all patients = 36 years.
Koronda Bcc+scc
Bee
1 5 2 3
_ 11
SCC
2 -_ 2
1 4 -
5
et aI??
Waider
Bee
+ see
-
BCC
---
see
Bee
-t see
-
-
-
-
3
-
et al.?j
3
1 1
cell carcinoma.
receiveconventionalimmunosuppressivetreatment, the basalcell carcinomato squamouscell carcinoma ratio was foundto be between1:4and 1:16.2$7$22In residentsof other countriesexposedto lessultraviolet radiation, the basal cell carcinomato squamous cell carcinomaratio is between1:0.36and 1:4.271 28 In other studiesin which cyclosporinewas given to patients who residedin countrieswith a northern latitude, the basal cell carcinomato squamouscell carcinomaratio was between9:0 and 1.4:l.*$g It is therefore important to considerthe country, latitude, amount of exposureto ultraviolet radiation, andthe follow-up time of the patientsin eachstudy. In studiesthat found an increasedincidenceof squamouscell carcinoma, the averagetime that elapsedbeforethe lesionappearedwaslong,between 30 and 86 months after surgery.4l7,22y2g,3oWhen the numberof basalcell carcinomaswasgreater,the averagetime that elapsedbeforetheir appearance was shorter(17.4 months).*We foundthat the time lapsefor basalcell carcinoma(25.3months)wasless than that for squamouscell carcinoma(36 months). Although the number of patients in our study is small, we noted that most neoplasmsappearedbetween the secondand third year after transplantation and that the ratio of basal cell carcinoma to squamouscell carcinomawas invertedas follow-up time progressed(Table III and Fig. 1). Nevertheless,onepatient had a squamouscell carcinomaon the lower lip 6 months after the transplant,which is early for the developmentof cancer“de nova,” seo ondaryto the immunosuppressionalone.We thereforeassumethat dysplasiamay havebeenpresentat the time of the transplant surgery. Most of our patientswith a skin neoplasmwere more than 50yearsof age(Table II). Thesefindings
reflectthoseof Olivari et al.* in heart transplantrecipients. However, in patients with kidney transplants, almost all were less than 50 years of age (Table V). In Spain, of all patients with skin neoplasms,the majority appearedafter the ageof 40 years;the averageage at the time squamouscell carcinomaappearedis higher still.23 Our findingselicit three questions:(1) What are the reasonsfor the elevatedincidenceof skin cancer in our study? (2) Why is the mean time of occurrence of basal cell carcinoma and squamouscell carcinoma different in this group of patients?(3) Why is the basal cell carcinoma to squamouscell carcinomaratio in our study differentthan the ratio in others studies conducted in countries with as much sunlight as Spain? Our findingsmay be causedby two factors.First, it hasbeendemonstratedexperimentallyin ratsthat cyclosporineadministration promotesthe survival and progressionof ultraviolet radiation-induced skin tumors via its capacity to enhancethe dominance of suppressorcell-controlled immune responses.2g Burnham et a1.31showedin rats that the carcinogenicpropertiesof ultraviolet radiation-inducedneoplasmswere inverselyproportionalto the expressionof major histocompatibilitycomplexclass I in lymphocytes.This growth decreasedwhen the cells were incubated with y-interferon. Furthermore,it is known that cyclosporineinhibits y-interferon synthesisby T cells.3oThis leadsto reduced expressionin the keratinocytesof intercellular adhesionmolecule-l 3oand thereforeto a reducedimmuneresponse.An absenceof intercellularadhesion molecule-l hasbeenfoundin tumor cellsin the case of basal cell carcinoma30T 32,33and squamouscell carcinoma,34> 35especiallywhenperitumoral mono-
Journal of the American Academy of Dermatology Volume 32, Number 3
nuclearcell infiltration is scarce,36 which occursin transplantrecipients.37Intercellular adhesionmolecule-l expressionreturnedafter the cellsin question had beenincubatedwith y-interferon.30a 34a 35The cancersmay ariseearlier in keratinocytesthat have previouslybeen damagedby ultraviolet radiation. For this reason,we believethat the main factorsin the appearanceof skin cancerin our patientswere skin type,previousexposureto ultraviolet radiation, and administration of cyclosporineand azathioprine. Azathioprine alsohasthe effect of promotingultraviolet radiation-initiated cancer, which has a greaterincidencethan that inducedby cyclosporine 38;this effect is directly related to the carcinogenic effect of azathioprine39and the levelsof oneof the metabolites of azathioprine (6-thioguanine).40 Moreover,it hasbeenshownthat azathioprine,unlike cyclosporine,inducesthe appearanceof keratotic and dysplasticlesions approximately 3 to 4 yearsafter it is takenandthat squamouslesionsmay later develop.39 This increasein the developmentof skin cancer,principally squamouscell carcinoma,is greaterin patientswho are sensitiveto UVB radiation and is mediated by tumor necrosisfactor.41 Thus in sometransplantrecipientsthe administration of certain immunosuppressantsincreasesthe progressionof previous dysplastic lesions or of incipient skin carcinoma,asperhapswasthe casein our patients. Second,many studiesindicateda high prevalence of human papillomavirusinfectionsin patientswho receivedtransplants.1,42-48However, more recent studies have not confirmed these findings.49The presenceof different human papillomavirus serotypesin actinic keratosisand in squamouscell carcinoma suggeststhat this virus may havean important role in the occurrenceof cancer.45-48 Different serotypeshave been described.28> 5o We therefore considerthe possibilitythat azathioprinemay play a greaterrole than was oncesupposedin inducing the appearanceof warts in photoexposedareas,as well as in later malignant development.32 Perhaps this is why a greaternumberof patientswho receive azathioprinetreatment have warts and squamous cell carcinoma.In supportof this theory, we have also observedthat when human papillomavirus is found in the squamouscell carcinoma of organ transplant recipients,multiple warts are found in photoexposedareas.16p 26-29The quantitative and qualitativedecreaseof Langerhanscellsin thesepa-
Espaiiu et al. 463 tients in areasexposedto the sun can also be of pathogenicimportance.51Therefore,patientswith organtransplantsshouldbe encouragedto usesunscreens.In our series,only 5 of the 14 patientswith skin cancerhad commonwarts (Table I). This incidencemay besosmall becauseall wartsweretreated promptly andall of our patientsusedsunblocksregularly. This may alsoexplainthe relativelylow incidenceof squamouscell carcinoma. Skin canceris moreproneto developin transplant recipientswho are homozygousfor HLA-DR, or who havethe HLA-A3 or HLA-B27 haplotype.In addition, those with one or two mismatches for HLA-B areat a relatively higherrisk of developing skin cancer.The presenceof HLA-Al 1 may play a preventiverole.31,38$39Nevertheless,someauthors havedescribedskin cancerin patientswith a kidney transplantwho expressedHLA-Al 1.52In our series we alsoencounteredonepatientwith the HLA-Al 1 haplotype who had basal cell carcinoma and squamouscell carcinoma(Table IV). However,the samplesize in our study is not large enoughfor us to draw anypreciseconclusions.Other authorshave found a statistically significant relationship for HLA-DR7 and HLA-B27 only.53We did not find a significant associationbetweenskin cancer and theseHLAs (Table V). REFERENCES
1. Abel EA. Cutaneousmanifestationsof immunosuppression in organ transplant recipients. J AM ACAD DERMATOL 1989;21:167-9. 2. Hardie IR, Strong RW, Hartley LCJ, et al. Skin cancerin Caucasianrenal allograft recipientsliving in a subtropical climate. Surgery 1980;87:177-83. 3. Koronda FC, Dehmel EM, Kahn G, et al. Cutaneouscomplications in immunosuppressedrenal homograft. JAMA 1974;229:419-24. 4. Gupta AK, Cardella CJ, Haberman HF. Cutaneous malignant neoplasmsin patients with renal transplants.Arch Dermatol 1986;122:1288-93. 5. Disler PB, MacPhail AP, Meyers AM, et al. Neoplasia after successfulrenal transplantation.Nephron 1981;29:11921. 6. Lanza RP, Cooper DKC, CassidyMJD, et al. Malignant neoplasmsoccurring after cardiactransplantation. JAMA 1983;249:1746-8. 7. Couetil JP, McGoldrick, Wallwork J, et al. Malignant tumors after heart transplantation. J Heart Lung Transplant 1990;9:622-6. 8. Olivari MT, Diekmann RA, Kubo SH, et al. Low incidence of neoplasiain heart and heart-lung transplant recipients receiving tripledrug immunosuppression.J Heart Lung Transplant 1990;9:618-1. 9. O’Connell BM, Abel EA, Nickoloff BJ,et al. Dermatologic complications following heart transplantation. J Heart Lung Transplant 1986;5:430-6.
464
Espaiiu et al.
10. McLelland J, RessA, Williams G, et al. The incidenceof immunosuppression-relatedskindiseasein long-term transplant patients. Transplantation 1988;46:871-4. 11. Ray JG, Hare DB, PedersonPD. NIAID manual of tissue typing techniques1979-1980. Washington, D.C.: Government Printing Office, 1979; DHEW publication no. (NIH) 80-545. 12. Van Rood JJ, Van Leeuwen A, Ploem JS. Simultaneous detection of the two cell populationsby two-colour fluorescenceand application to the recognition of B-celldeterminants. Nature 1976;262:795-7. 13. Bouwes-BavinckJN, Koote AMM, Van der Woude FJ, et al. HLA-Al 1 associatedresistanceto skin cancer in renal transplant recipients.N Engl J Med 1990;323:1350. 14. Bouwes-BavinckJN, Vermeer BJ, Van der Woude FJ, et al. Relation between skin cancer and HLA antigens in re nal transplant recipients.N Engl J Med 1991;325:843-8. 15. BouwesBavinckJN, Kootte AMM, Van der Woude FJ, et al. On a possibleprotectiveeffect of HLA-Al 1 against skin cancer and keratotic skin lesionsin renal transplant recipients. J Invest Dermatol 1991;97:269-72. 16. Kahn HA. Longitudinal studies: person-year. In: Kahn HA, ed. An introduction to epidemiologicmethods. New York: Oxford University Press,1983:146-60. 17. Kahn HA. Longitudinal studies:life tables. In: Kahn HA, ed. An introduction to epidemiologicmethods. New York: Oxford University Press,1983:121-45. 18. Penn I, First MR. Development and incidenceof cancer following cyclosporine therapy. Transplant Proc 1986; 17:210-5. 19. Penn I. Cancersfollowing cyclosporinetherapy. Transplantation 1987;43:32-5. 20. Blohme I, Larko 0. No difference in skin cancerincidence with or without cyclosporine:a 5-year perspective.Transplant Proc 1992243 13. 21. Bergfeld WF, Roenigk HH. Cutaneous complicationsof immunosuppressivetherapy: a review of 215 renal transplant patients. Cutis 1978;22:169-72. 22. Walder BK, Jeremy D, Charlesworth JA, et al. The skin and immunosuppression. Australas J Dermatol 1976; 17:94-7. 23. Zubiri A, Berrino F, de1Moral A, et al. Estadisticade 10s cancerescutaneosepitelialesen 10spaiseseuropeosde lengua latina. Actas Dermo-Sif 1981;5-6:241-50. 24. Krupp P, Monka C. Side-effectprofile of cyclosporineA in patientstreated for psoriasis.Br J Dermatoll990; 122(suppl 36):47-56. 25. Von Graffenried B, Friend D, Shand N, et al. Cyclosporin A (Sandiiune (R)) in autoimmune disorders. In: Thomson AW, ed. Sandimmune (ciclosporin):mode of action and clinical application. Lancaster, U.K.: Kluwer Academic, 1989:213-51. 26. Hartevelt MM, BouwesBavinckJN, Kootte AMM, et al. Incidence of skin cancerafter renal transplantation in the Netherlands. Transplantation 1990;49:506-9. 27. Walder BK, Robertson MR, Jeremy D. Skin cancer and immunosuppression.Lancet 1971;2:1282-3. 28. Euvrard S, Chardonnet Y, Dureau G, et al. Human papillomavirus type l-associated squamouscell carcinoma in a heart transplant recipient. Arch Dermatol1991;127:55964. 29. Servilla K, Burnham DK, Daynes RA. Ability of cyclosporin to promote the growth of transplanted ultraviolet radiation-induced tumours in mice. Transplantation 1987; 44:291-5. 30. Nickoloff BJ, Grifhths CEM, Barker JNWN. The role of
Journal of the American Academy of Dermatology March 1995
adhesionmolecules,chemotactic factors, and cytokinesin inflammatory and neoplasticskin,J Invest Dermatol1990; 94(suppl):151S-7s. 31. Burnham DK, Gahring LC, Daynes RA. Clonal origin of tumors induced by ultraviolet radiation. J Nat1 Cancer Inst 1986;76:151-8. 32. Barker JNWN, Allen MH, MacDonald DM. Distribution of intercellular adhesionmolecule-l (ICAM-1) and lymphocyte function associatedantigen-l (LFA-1) in epiderma1tumours [Abstract]. J Invest Dermatol 1989;93:294. 33. Taylor RS, Grifbths CEM, Brown MD, et al. Constitutive absenceand interferon-y-induced expressionof adhesion moleculesin basalcell carcinoma.J AM ACADDERMATOL 1990;22:721-6. 34. Scher RL, Koch WM, Richtsmeier WJ. Induction of intercellular adhesion molecule (ICAM- 1) on squamous cell carcinoma by interferon gamma. Arch Otolaryngol Head Neck Surg 1993;119:432-8. 35. Yago T, Tukuda M, Mocimatu I, et al. The expresionof ICAM- on head and neck squamouscellcarcinoma.Nippon Jibiinkoka Gakkai Kaiho 1993;96:827-32. 36. Barker JNWN, Allen MH, MacDonald DM. Distribution of intercellular adhesionmolecule1 (ICAM-1) and lymphocyte-function-associatedantigen-l (LFA-1) in epiderma1tumours. Clin Exp Dermatol 1990;15:331-4. 37. Hoyo E, Kanitakis J, Euvrard S, et al. Proliferation characteristicsof cutaneoussquamouscell carcinomasdeveloping in organ graft recipients. Arch Dermatol 1993; 129:324-7. 38. Kelly GE, Meike W, Sheil AGR. Effects of immunosuppressivetherapy on the induction of skin tumours by ultraviolet irradiation in hairlessmice. Transplantation 1987; 44429-4. 39. Taylor AEM, ShusterS. Skin cancerafter renal transplantation: the causalrole of azathioprine.Acta Derm Venereol (Stockh) 1992;72:115-9. 40. Lennard L, Thomas S, Harrington CI, et al. Skin cancerin renal transplant recipientsis associatedwith increasedconcentrationsof 6-thioguaninenucleotidein red blood cells.Br J Dermatol 1985;113:723-9. 41. Streilein JW. Immunogenetic factors in skin cancer. N Engl J Med 1991;325:884-6. 42. Goldstein GD, Gollub S, Gill B. Cutaneouscomplications of heart transplantation. J Heart Lung Transplant 1986; 5:143-7. 43. Rudlinger R, Smith IW, Bunney MH, et al. Human papillomavirusinfections in a group of renal transplant recipients. Br J Dermatol 1986;115:681-92. 44. Van der Leest RJ, Zachow KR, Ostrow RS, et al. Human papillomavirusheterogeneity in 36 renal transplant recip ients. Arch Dermatol 1987;123:354-7. 45. Barr BB, Benton EC, McLaren K, et al. Human papillomavirus infection and skin cancer in renal allograft recipients. Lancet 1989;1:124-8. 46. Pfister H, GrossG, Hagedorn M. Characterization of human papillomavirus 3 in warts of renal allograft patient. J Invest Dermatol 1979;73:349-53. 47. Lutzner M, Croissant0, DucasseMF, et al. A potentially oncogenichuman papillomavirus (HPV-5) found in two renal allograft recipients.J Invest Dermatoll980;75:353-6. 48. Lutzner MA, Orth G, Dutronquay V, et al. Detection of human papillomavirus type 5 DNA in skin cancersof an immunosuppressedrenal allograft recipients.Lancet 1983; 21422-4. 49. Smith SE, Davis IC, Leshm B, et al. Absenceof human papillomavirusin squamouscell carcinomasof nongenital
Journal of the American Academy of Dermatology Volume 32, Number 3
Wittenberg,Oursler, and Peters
skin from immunccompromisedrenal transplant patients. Arch Dermatol 1993;129:1585-8. 50. Demetrick DJ, Inoue M, Lester WM, et al. Human papillomavirus type 16 associatedwith oral squamouscarcinoma in a cardiac transplant recipient. Cancer 1990; 66:1726-31. 51. Sontheimer RD, BergstresserPR, Gammas P, et al. Perturbation of epidermal Langerhans cells in immunosup
pressedhuman renal allograft recipients.Transplantation 1984;37:168-74. 52. McGregor JM, Reddi G, MacDonald D, et al. HLA-Al 1 in renal allograft recipientswith skin cancer.J Invest Dermatol 1991;98:261-2. 53. CzarneckiD, Watkins F, Leahy S, et al. Skin cancersand HLA frequenciesin renal transplant recipients.Dermatology 1992;185:9-11.
Secondary amyloidosis complicating psoriasis Gregory P. Wittenberg,
MD, Judith R. Oursler, MD, and Margot
S. Peters, MD
Rochester, Minnesota Background: Amyloidosis is a heterogeneous group of disorders characterized by extracellular deposition of fibrillar protein. Secondary amyloidosis occurs in patients with chronic infectious or inflammatory processes. Only 18 cases of secondary amyloidosis complicating
psoriasis have been reported. Objective: We characterized secondary amyloidosis complicating psoriasis. Methods: We reviewed all cases of coexisting psoriasis and amyloidosis seen at the Mayo Clinic from 1950 to 1992. The clinical characteristics were summarized, and the literature was reviewed. Results: There were 28 casesof coexistent disease, and in five of these psoriasis was the only inflammatory condition preceding the development of secondary amyloidosis. Histopathologic confirmation with Congo red staining was available in four cases, and immunohistochemical confirmation of the characteristic amyloid A subtype was performed in two.
Conhsiont Follow-up of four patients supports the view that amyloidosis associated with psoriasis is an aggressive disease that may be fatal. (J AM ACAD DERMATOL 1995;32:465-8.) Amyloidosis is defined as the extracellular deposition of amyloid, a protein composed of fine, nonbranching fibrils that aggregate into P-pleated sheets. Amyloidosis has been divided into subtypes on the basis of biochemical and clinical characteristics. Some clinical subtypes include primary amyloidosis, heredofamilial amyloidosis, local amyloidosis, senile amyloidosis, and amyloidosis associated with long-term hemodialysis. The two most common biochemical subtypes are amyloid A (AA), found in secondary and some familial forms of the disorder, and amyloid L, found in primary and multiple myeloma-associated amyloidosis. ’ From the Department dation. Accepted Reprint Mayo Copyright
of Dermatology,
for publication
Nov.
Mayo
Clinic
Foun-
7, 1994.
requests: M. S. Peters, MD, Department Clinic, 200 First St. SW, Rochester, MN @ 1995 by the American
0190-9622/95 $3.00+0
and Mayo
Academy
16/l/61867
of Dermatology, 55905.
of Dermatology,
Inc.
Secondary amyloidosis is commonly associated with a variety of chronic inflammatory disorders such as rheumatoid arthritis, ankylosing spondylitis, and osteomyelitis. It has been only rarely reported to be associated with psoriasis. AA can be isolated from secondary amyloid. This is a unique, insoluble protein with a molecular weight of approximately 8.5 kd. It is made up of 76 amino acid residues arranged in a single chain. The putative precursor to AA is an antigenically related protein, serum AA (SAA). This is a normal soluble serum component that behaves as an acute-phase reactant. SAA associates with the high-density lipoprotein-3 subclass of serum lipoproteins and is therefore sometimes referred to as apo-SAA. It is synthesized and released by hepatocytes under the control of interleukin a. The precise regulation of the conversion of SAA to the insoluble AA protein of amyloidosis is not understood. However, the induction of AA amyloidosis is thought to be a multifactorial process influenced by 465
AND METHODS
We determined the presence of skin cancer (basal cell carcinoma and squamous cell carcinoma) in 92 of 111 patients who had undergone heart transplantation from January 1984 to December 1993. Like others authors,7 From the Departments of Dermatologya University Clinic of Navarra, School Accepted
for publication
and Cardiovascular of Medicine.
Surgery,b
Sept. 12, 1994
Reprint requests: Agustin Espafia, MD, Department of Dermatology, University Clinic of Navarra, School of Medicine, P.O. Box 192, 31080 Pamplona, Spain. Copyright
@ 1994 by the American
0190-9622/95/$3.00+0
458
16/l/60703
Academy
of Dermatology,
Inc.
we did not include 16 patients who died within 6 months after undergoing transplantation and three patients with a follow-up period of lessthan 3 months after undergoing transplantation. All patients came from the northern and eastern regions of Spain (latitude 40” to 45’ N). Patients ranged in age from 2 months to 69 years (average, 49.5 years). Seven patients were lessthan 15 years of age. The average follow-up period of the 111 patients was 43.3 months (range, 1 to 114 months). Twenty-two patients died. Data from 1984 to 1990 were obtained from medical records. Two dermatologists examined the patients every 6 months, concentrating particularly on the areas of their bodies exposed to sunlight. All suspicious lesions were biopsied. Simple excision was used to treat all patients with a basal cell carcinoma or squamous cell carcinoma. In one patient with squamous cell carcinoma of the lower lip, cervical lymphadenectomy and radiotherapy were also performed. The patient died of generalized metastases. Photoprotective measures were initiated in all patients. The skin type I to VI was assessedaccording to the standard classification9 Details of exposure to ultraviolet radiation in each patient were obtained from the medical
Journal of the American Academy of Dermatology Volume 32, Number 3
Table I. Epidemiologic skin
cancer
459
data
Patient No.
Time of appearance (mo since- transplant)
%e m
BCC BCC BCC BCC BCC
69 38 38 68 55
see see see see see see
6 7 8 9 10 11
64 61 53 53 67 62
BCC + SCC
12
57
BCC + SCC
13
57
BCC + SCC
14
50
BCC, Basal cell carcinoma; *Basal cell carcinoma. fSquamous cell carcinoma.
Espaiia et al.
SCC, squamous
7 36 11 20 10 21 26 26 13 6 72 53 22 24 24 36” 36” 367 26* 41* 26t 31”1‘ 60* 50t 58 547
Skill Warts
-
+ + +
-
type
II III III III II
-
II II II II II II
+
II
-
II
+
II
-
LOCAiOll
Face Face Back Face Face Ear Forehead Scalp Arm Lower lip Lower lip Ear lobe Lower lip Forehead Forehead Face Back Face Face Face Face Face Scalp Lower lip Ear lobe Neck
Profession
Farmer Civil servant Civil servant Businessman Businessman
Civil servant Driver Civil servant Distributor Farmer Civil servant Businessman Bricklayer
Farmer
cell carcinoma
records, following the criteria established by McLelland et al.‘O HLA-A and HLA-B typing was performed in all patients before transplantation, and HLA-DR was determined in 95 patients. t 1,t2 The degree of mismatching was assessedby counting the antigens present in the donor but absent in the recipient. We performed correlations only with HLA-A3, HLA-Al 1, HLA-B27, HLA-DR homozygosity, and HLA-B mismatching, because these are the only haplotypes that have been related to the occurrence of neoplasms in transplant recipients.13-t5 The incidence of skin cancer development (basal cell carcinoma and squamous cell carcinoma) after transplantation per 1000 person-years was computed16 for the overall follow-up period and for the periods of 0 to 12,13 to 24, 25 to 36, 37 to 48, 49 to 60, 61 to 72, and longer than 72 months after transplantation. We also computed the cumulative incidence of skin cancer by life-table analysis.l7 For statistical purposes, data for a patient were included in a given period if he or she had been observed for that entire period or at least for two thirds of that period. When this was not the case, the data for this patient
were included in the previous period, if this situation OG curred during the first year, the data were not taken into account. Fisher’s test was used to assessfor statistically significant relationships between HLA matching and the presence of skin cancer. The following immunosuppressive protocols were used for all patients in each period: 1984 to 1986: Cyclosporine (6 mg/kg before surgery and, according to blood levels, between 200 and 300 rig/ml after surgery); 6-methylprednisolone (500 mg intravenously intraoperatively, and 125 mg three times a day intravenously for 3 days); and oral prednisone (1 mg/kg/day after surgery, reduced to 0.2 mg/kg/day at the end of the fourth week). In patients with renal insufficiency after surgery (creatinine > 1.5), cyclosporine was replaced by equine antithymocytic y-globulin (10 to 20 mg/kg/day) for 7 days. 1986 to 1989: Azathioprine (2 mg/kg intravenously before surgery and 1.5 to 2 mg/kg/day after surgery) and corticosteroid administration as used in patients from 1984 to 1986. Cyclosporine was administered on the third day after surgery and afterward according to blood lev-
460
Journal of the American Academy of Dermatology March 1995
Espaiia et al. 50
Table II. Age/skin cancerdistribution
1
Skin cancer Age w*
BCC
20-29 30-39 40-49 50-59 60-70 Total
1 1 2 5
SCC
-
-
2 4 is
3 5
KC, Basal cell carcinoma; SCC, squamous *Average age of all patients = 49.5 years. 0
2
YEARS
AFTER
4
6
BCC + SCC
cell carcinoma.
a
TRANSPLANTATION
Fig. 1. Cumulative
incidence of total skin cancer, squamous cellcarcinoma,and basal cell carcinoma. HoZlow square, Basal cell carcinoma; solid square, total skin cancer;diamond, squamouscellcarcinoma.
Table III. Relation betweenfollow-up time and type of skin cancer Posttransplant period No. of No. of patients No. of patients with see (mo) I patients I with BCC
o-12 13-24 25-36 37-48 49-60 61-72 >72
els. The same method as that used from 1984 to 1986 was followed with respect to patients with renal insufficiency. I989 to 1993: The protocol was the same as that used for patientsfrom 1986to 1989;however, in patientswith
creatininegreaterthan1S, cyclosporine wassubstituted for OKT3 (5 mg/dayintravenously for 12days).
KC,
Basal cell carcinoma;
92 85 65 51 34 26 16 SCC, squamous
3 2 6 1 1 0 0
1 4 3 0 ‘4 1 0 cell carcinoma.
RESULTS
In 14 patients (15.2%), 26 skin cancerswere found. Five patients had basal cell carcinoma, six had squamouscell carcinoma,and three had both basalcell carcinomaand squamouscell carcinoma (Table I). The basalcell carcinomato squamouscell carcinomaratio was 1:1.5 for the first malignancy. When all secondand subsequentskin cancerswere included,the basalcell carcinomato squamouscell carcinoma ratio was 1:1.3. In four of the patients with squamouscell carcinoma,the lesiondeveloped on the lower lip (at the vermilion border exclusively). All patientshad skin type II or III (Table I). In the first phaseof data gatheringwe identifiedthree patients with skin cancer (patients 2, 6, and 12), whereas11 patientswith skin cancerwere detected in the prospectivefollow-up (patients 1,3,4,5,7,8, 9, 10, 11, 13,and 14) (Table I). Table II showsthe agedistribution of the 14 patientsin whom one or more skin neoplasmsoccurred. Twelve patients (11.5%) had common warts, which appearedbetween 2 and 55 months after transplantation(mean,21.7months).All wartswere
locatedin areasexposedto ultraviolet radiation,and they respondedto cryotherapy.Five of these12 patients had an associatedneoplasm or neoplasms (patient 2, basalcell carcinoma;patient4, basalcell carcinoma;patient 5, basalcell carcinoma;patient 12, basal cell carcinoma and squamouscell carcinoma; and patient 14, basal cell carcinoma and squamouscell carcinoma) (Table I). Overall, skin cancerappearedan averageof 31.5 months after transplantation;basalcell carcinoma appearedat an averageof 25.3 months, and the squamouscell carcinoma appearedat 36 months. Most cancersappearedin the first 4 years after transplantation(Table III). The overallincidenceof both typesof skin cancer was45.3casesper 1000posttransplantperson-years: the incidencewas25.88for basalcell carcinomaand 29.12for squamouscell carcinoma.The overallrisk of developingskin cancerincreasedfrom a cumulative incidenceof 4.3% 1year after transplantationto 43.8% 7 years after transplantation.For basal cell carcinomathe incidenceswere 3.26% after 1 year
Journal of the American Academy of Dermatology Volume 32, Number 3
Espahaet al. 461
Table IV. HLA distribution Type of skin cancer HLA
study
No. of patients with skin cancer
SCC
1 4
4 2
2
19 72
0 4
0 7
1
2
6 85
0 4
0 7
0 3
90
0
2
1
4
5
2
0
0
3
4 3
0 2 P
A3 (NS) Positive 6 Negative 8 All (NS) Positive 1 Negative 13 B27 (NS) Positive 0 Negative 14 DR homozygosity (NS) Positive 3 Negative 11 No. mismatches for HLA-B (NS) 0 0 1 9
2 KC,
Basal cell carcinoma;
5 SCC, squamous
1 cell carcinoma;
BCC + SCC
No. patients without skin cancer
BCC
1
1
38 45 1 18
72
NS, not significant.
and 16.24%after 7 years, and for squamouscell carcinomathe incidenceswere 1.08%after 1 year and 27.55%after 7 years (Fig. 1). The HLA distributions are shownin Table IV. We did not find a significant associationbetween skincancerandtheseHLAs. We did not observeany casesof Kaposi’ssarcomaor malignant melanoma. DISCUSSION Skin canceraccountsfor mostof the neoplasmsin transplantrecipients.2~9l1*, 19Overall,the frequency of skin cancerin patientstaking conventionalimmunosuppressiveagentsis 40%; the frequency is only 15%whencyclosporineis administered.l8 Nevertheless,Blohme and Larko2’ found no differences in the frequencyof skin cancerin patientsreceiving azathioprine,cyclosporine,or combinedtherapy.In other studies with patients who did not receive treatment with cyclosporine,the frequencyof skin cancerrangedfrom 0.5%to 15.6%.21> 22In patients who receivedtreatment with cyclosporine,the frequencyrangedbetween0% to 11%.21 9 Our results, which indicate a frequencyof 15.2%,are in agreement with thesefindings. In our area of Spain the frequencyof skin neoplasmsin the population at large is 1.8%F3approximately one seventhof the figure obtainedin this series. Little information is available about cutaneous neoplasmsin patientsnot receivingtransplantswho
were treated with cyclosporinefor other diseases. Krupp and Monka observeda frequencyof 0.7%, with a basalcell carcinomato squamouscell carcinoma ratio of 1:6.However,all of their patientshad previouslybeentreatedwith PUVA andmethotrexate. These authors believedthat the cancersthey observedwereunrelatedto cyclosporine,becausethe patientshad beentaking this drug for only a short time. No referencehasbeenmadeto skin cancerin the context of other autoirnmune diseases.25 In the statistical analysis of our series,an increasedcumulative incidencewas noted (Fig. 1). Our frequencyis slightly greaterthan that reported by Hardie et a1.,2who found a cumulative risk of 38%after 8 years.In our seriesthe degreeof risk was 43.8% after 7 years. The reason for this high
incidencecouldbethat both studieswerecarriedout in countrieswith high levelsof ultraviolet radiation. In othercountrieswith lesssolarradiation,this risk doesnot reach 10% after 7 years.26Similarly, we found an incidenceof 45.3 casesper 1000personyears,which resemblesthat reportedin othercountries with high levels of ultraviolet radiation (43 cases per 1000 person-years)and exceedsthat encounteredin countries with less sunlight (nine casesper 1000person-years). The basalcell carcinomato squamouscell carcinoma ratio in Latin countrieshasbeenreportedas 2.1 . .23 Among residents of sunny countries who
462
Journal of the American Academy of Dermatology March 1995
Espaiiu et al.
Table V. Age/skin cancerdistribution in patientswith classicimmunosuppressivetreatment Hardie w
20-30 31-40 41-50 51-60 >61 Total
BCC
see
-
-
1 3 _ 4
et al.**
4 7 -
_ 11
BCC, Basal cell carcinoma; XC, squamous *Average age of all patients = 30 years. TAverage age of all patients = 36 years. SAverage age of all patients = 36 years.
Koronda Bcc+scc
Bee
1 5 2 3
_ 11
SCC
2 -_ 2
1 4 -
5
et aI??
Waider
Bee
+ see
-
BCC
---
see
Bee
-t see
-
-
-
-
3
-
et al.?j
3
1 1
cell carcinoma.
receiveconventionalimmunosuppressivetreatment, the basalcell carcinomato squamouscell carcinoma ratio was foundto be between1:4and 1:16.2$7$22In residentsof other countriesexposedto lessultraviolet radiation, the basal cell carcinomato squamous cell carcinomaratio is between1:0.36and 1:4.271 28 In other studiesin which cyclosporinewas given to patients who residedin countrieswith a northern latitude, the basal cell carcinomato squamouscell carcinomaratio was between9:0 and 1.4:l.*$g It is therefore important to considerthe country, latitude, amount of exposureto ultraviolet radiation, andthe follow-up time of the patientsin eachstudy. In studiesthat found an increasedincidenceof squamouscell carcinoma, the averagetime that elapsedbeforethe lesionappearedwaslong,between 30 and 86 months after surgery.4l7,22y2g,3oWhen the numberof basalcell carcinomaswasgreater,the averagetime that elapsedbeforetheir appearance was shorter(17.4 months).*We foundthat the time lapsefor basalcell carcinoma(25.3months)wasless than that for squamouscell carcinoma(36 months). Although the number of patients in our study is small, we noted that most neoplasmsappearedbetween the secondand third year after transplantation and that the ratio of basal cell carcinoma to squamouscell carcinomawas invertedas follow-up time progressed(Table III and Fig. 1). Nevertheless,onepatient had a squamouscell carcinomaon the lower lip 6 months after the transplant,which is early for the developmentof cancer“de nova,” seo ondaryto the immunosuppressionalone.We thereforeassumethat dysplasiamay havebeenpresentat the time of the transplant surgery. Most of our patientswith a skin neoplasmwere more than 50yearsof age(Table II). Thesefindings
reflectthoseof Olivari et al.* in heart transplantrecipients. However, in patients with kidney transplants, almost all were less than 50 years of age (Table V). In Spain, of all patients with skin neoplasms,the majority appearedafter the ageof 40 years;the averageage at the time squamouscell carcinomaappearedis higher still.23 Our findingselicit three questions:(1) What are the reasonsfor the elevatedincidenceof skin cancer in our study? (2) Why is the mean time of occurrence of basal cell carcinoma and squamouscell carcinoma different in this group of patients?(3) Why is the basal cell carcinoma to squamouscell carcinomaratio in our study differentthan the ratio in others studies conducted in countries with as much sunlight as Spain? Our findingsmay be causedby two factors.First, it hasbeendemonstratedexperimentallyin ratsthat cyclosporineadministration promotesthe survival and progressionof ultraviolet radiation-induced skin tumors via its capacity to enhancethe dominance of suppressorcell-controlled immune responses.2g Burnham et a1.31showedin rats that the carcinogenicpropertiesof ultraviolet radiation-inducedneoplasmswere inverselyproportionalto the expressionof major histocompatibilitycomplexclass I in lymphocytes.This growth decreasedwhen the cells were incubated with y-interferon. Furthermore,it is known that cyclosporineinhibits y-interferon synthesisby T cells.3oThis leadsto reduced expressionin the keratinocytesof intercellular adhesionmolecule-l 3oand thereforeto a reducedimmuneresponse.An absenceof intercellularadhesion molecule-l hasbeenfoundin tumor cellsin the case of basal cell carcinoma30T 32,33and squamouscell carcinoma,34> 35especiallywhenperitumoral mono-
Journal of the American Academy of Dermatology Volume 32, Number 3
nuclearcell infiltration is scarce,36 which occursin transplantrecipients.37Intercellular adhesionmolecule-l expressionreturnedafter the cellsin question had beenincubatedwith y-interferon.30a 34a 35The cancersmay ariseearlier in keratinocytesthat have previouslybeen damagedby ultraviolet radiation. For this reason,we believethat the main factorsin the appearanceof skin cancerin our patientswere skin type,previousexposureto ultraviolet radiation, and administration of cyclosporineand azathioprine. Azathioprine alsohasthe effect of promotingultraviolet radiation-initiated cancer, which has a greaterincidencethan that inducedby cyclosporine 38;this effect is directly related to the carcinogenic effect of azathioprine39and the levelsof oneof the metabolites of azathioprine (6-thioguanine).40 Moreover,it hasbeenshownthat azathioprine,unlike cyclosporine,inducesthe appearanceof keratotic and dysplasticlesions approximately 3 to 4 yearsafter it is takenandthat squamouslesionsmay later develop.39 This increasein the developmentof skin cancer,principally squamouscell carcinoma,is greaterin patientswho are sensitiveto UVB radiation and is mediated by tumor necrosisfactor.41 Thus in sometransplantrecipientsthe administration of certain immunosuppressantsincreasesthe progressionof previous dysplastic lesions or of incipient skin carcinoma,asperhapswasthe casein our patients. Second,many studiesindicateda high prevalence of human papillomavirusinfectionsin patientswho receivedtransplants.1,42-48However, more recent studies have not confirmed these findings.49The presenceof different human papillomavirus serotypesin actinic keratosisand in squamouscell carcinoma suggeststhat this virus may havean important role in the occurrenceof cancer.45-48 Different serotypeshave been described.28> 5o We therefore considerthe possibilitythat azathioprinemay play a greaterrole than was oncesupposedin inducing the appearanceof warts in photoexposedareas,as well as in later malignant development.32 Perhaps this is why a greaternumberof patientswho receive azathioprinetreatment have warts and squamous cell carcinoma.In supportof this theory, we have also observedthat when human papillomavirus is found in the squamouscell carcinoma of organ transplant recipients,multiple warts are found in photoexposedareas.16p 26-29The quantitative and qualitativedecreaseof Langerhanscellsin thesepa-
Espaiiu et al. 463 tients in areasexposedto the sun can also be of pathogenicimportance.51Therefore,patientswith organtransplantsshouldbe encouragedto usesunscreens.In our series,only 5 of the 14 patientswith skin cancerhad commonwarts (Table I). This incidencemay besosmall becauseall wartsweretreated promptly andall of our patientsusedsunblocksregularly. This may alsoexplainthe relativelylow incidenceof squamouscell carcinoma. Skin canceris moreproneto developin transplant recipientswho are homozygousfor HLA-DR, or who havethe HLA-A3 or HLA-B27 haplotype.In addition, those with one or two mismatches for HLA-B areat a relatively higherrisk of developing skin cancer.The presenceof HLA-Al 1 may play a preventiverole.31,38$39Nevertheless,someauthors havedescribedskin cancerin patientswith a kidney transplantwho expressedHLA-Al 1.52In our series we alsoencounteredonepatientwith the HLA-Al 1 haplotype who had basal cell carcinoma and squamouscell carcinoma(Table IV). However,the samplesize in our study is not large enoughfor us to draw anypreciseconclusions.Other authorshave found a statistically significant relationship for HLA-DR7 and HLA-B27 only.53We did not find a significant associationbetweenskin cancer and theseHLAs (Table V). REFERENCES
1. Abel EA. Cutaneousmanifestationsof immunosuppression in organ transplant recipients. J AM ACAD DERMATOL 1989;21:167-9. 2. Hardie IR, Strong RW, Hartley LCJ, et al. Skin cancerin Caucasianrenal allograft recipientsliving in a subtropical climate. Surgery 1980;87:177-83. 3. Koronda FC, Dehmel EM, Kahn G, et al. Cutaneouscomplications in immunosuppressedrenal homograft. JAMA 1974;229:419-24. 4. Gupta AK, Cardella CJ, Haberman HF. Cutaneous malignant neoplasmsin patients with renal transplants.Arch Dermatol 1986;122:1288-93. 5. Disler PB, MacPhail AP, Meyers AM, et al. Neoplasia after successfulrenal transplantation.Nephron 1981;29:11921. 6. Lanza RP, Cooper DKC, CassidyMJD, et al. Malignant neoplasmsoccurring after cardiactransplantation. JAMA 1983;249:1746-8. 7. Couetil JP, McGoldrick, Wallwork J, et al. Malignant tumors after heart transplantation. J Heart Lung Transplant 1990;9:622-6. 8. Olivari MT, Diekmann RA, Kubo SH, et al. Low incidence of neoplasiain heart and heart-lung transplant recipients receiving tripledrug immunosuppression.J Heart Lung Transplant 1990;9:618-1. 9. O’Connell BM, Abel EA, Nickoloff BJ,et al. Dermatologic complications following heart transplantation. J Heart Lung Transplant 1986;5:430-6.
464
Espaiiu et al.
10. McLelland J, RessA, Williams G, et al. The incidenceof immunosuppression-relatedskindiseasein long-term transplant patients. Transplantation 1988;46:871-4. 11. Ray JG, Hare DB, PedersonPD. NIAID manual of tissue typing techniques1979-1980. Washington, D.C.: Government Printing Office, 1979; DHEW publication no. (NIH) 80-545. 12. Van Rood JJ, Van Leeuwen A, Ploem JS. Simultaneous detection of the two cell populationsby two-colour fluorescenceand application to the recognition of B-celldeterminants. Nature 1976;262:795-7. 13. Bouwes-BavinckJN, Koote AMM, Van der Woude FJ, et al. HLA-Al 1 associatedresistanceto skin cancer in renal transplant recipients.N Engl J Med 1990;323:1350. 14. Bouwes-BavinckJN, Vermeer BJ, Van der Woude FJ, et al. Relation between skin cancer and HLA antigens in re nal transplant recipients.N Engl J Med 1991;325:843-8. 15. BouwesBavinckJN, Kootte AMM, Van der Woude FJ, et al. On a possibleprotectiveeffect of HLA-Al 1 against skin cancer and keratotic skin lesionsin renal transplant recipients. J Invest Dermatol 1991;97:269-72. 16. Kahn HA. Longitudinal studies: person-year. In: Kahn HA, ed. An introduction to epidemiologicmethods. New York: Oxford University Press,1983:146-60. 17. Kahn HA. Longitudinal studies:life tables. In: Kahn HA, ed. An introduction to epidemiologicmethods. New York: Oxford University Press,1983:121-45. 18. Penn I, First MR. Development and incidenceof cancer following cyclosporine therapy. Transplant Proc 1986; 17:210-5. 19. Penn I. Cancersfollowing cyclosporinetherapy. Transplantation 1987;43:32-5. 20. Blohme I, Larko 0. No difference in skin cancerincidence with or without cyclosporine:a 5-year perspective.Transplant Proc 1992243 13. 21. Bergfeld WF, Roenigk HH. Cutaneous complicationsof immunosuppressivetherapy: a review of 215 renal transplant patients. Cutis 1978;22:169-72. 22. Walder BK, Jeremy D, Charlesworth JA, et al. The skin and immunosuppression. Australas J Dermatol 1976; 17:94-7. 23. Zubiri A, Berrino F, de1Moral A, et al. Estadisticade 10s cancerescutaneosepitelialesen 10spaiseseuropeosde lengua latina. Actas Dermo-Sif 1981;5-6:241-50. 24. Krupp P, Monka C. Side-effectprofile of cyclosporineA in patientstreated for psoriasis.Br J Dermatoll990; 122(suppl 36):47-56. 25. Von Graffenried B, Friend D, Shand N, et al. Cyclosporin A (Sandiiune (R)) in autoimmune disorders. In: Thomson AW, ed. Sandimmune (ciclosporin):mode of action and clinical application. Lancaster, U.K.: Kluwer Academic, 1989:213-51. 26. Hartevelt MM, BouwesBavinckJN, Kootte AMM, et al. Incidence of skin cancerafter renal transplantation in the Netherlands. Transplantation 1990;49:506-9. 27. Walder BK, Robertson MR, Jeremy D. Skin cancer and immunosuppression.Lancet 1971;2:1282-3. 28. Euvrard S, Chardonnet Y, Dureau G, et al. Human papillomavirus type l-associated squamouscell carcinoma in a heart transplant recipient. Arch Dermatol1991;127:55964. 29. Servilla K, Burnham DK, Daynes RA. Ability of cyclosporin to promote the growth of transplanted ultraviolet radiation-induced tumours in mice. Transplantation 1987; 44:291-5. 30. Nickoloff BJ, Grifhths CEM, Barker JNWN. The role of
Journal of the American Academy of Dermatology March 1995
adhesionmolecules,chemotactic factors, and cytokinesin inflammatory and neoplasticskin,J Invest Dermatol1990; 94(suppl):151S-7s. 31. Burnham DK, Gahring LC, Daynes RA. Clonal origin of tumors induced by ultraviolet radiation. J Nat1 Cancer Inst 1986;76:151-8. 32. Barker JNWN, Allen MH, MacDonald DM. Distribution of intercellular adhesionmolecule-l (ICAM-1) and lymphocyte function associatedantigen-l (LFA-1) in epiderma1tumours [Abstract]. J Invest Dermatol 1989;93:294. 33. Taylor RS, Grifbths CEM, Brown MD, et al. Constitutive absenceand interferon-y-induced expressionof adhesion moleculesin basalcell carcinoma.J AM ACADDERMATOL 1990;22:721-6. 34. Scher RL, Koch WM, Richtsmeier WJ. Induction of intercellular adhesion molecule (ICAM- 1) on squamous cell carcinoma by interferon gamma. Arch Otolaryngol Head Neck Surg 1993;119:432-8. 35. Yago T, Tukuda M, Mocimatu I, et al. The expresionof ICAM- on head and neck squamouscellcarcinoma.Nippon Jibiinkoka Gakkai Kaiho 1993;96:827-32. 36. Barker JNWN, Allen MH, MacDonald DM. Distribution of intercellular adhesionmolecule1 (ICAM-1) and lymphocyte-function-associatedantigen-l (LFA-1) in epiderma1tumours. Clin Exp Dermatol 1990;15:331-4. 37. Hoyo E, Kanitakis J, Euvrard S, et al. Proliferation characteristicsof cutaneoussquamouscell carcinomasdeveloping in organ graft recipients. Arch Dermatol 1993; 129:324-7. 38. Kelly GE, Meike W, Sheil AGR. Effects of immunosuppressivetherapy on the induction of skin tumours by ultraviolet irradiation in hairlessmice. Transplantation 1987; 44429-4. 39. Taylor AEM, ShusterS. Skin cancerafter renal transplantation: the causalrole of azathioprine.Acta Derm Venereol (Stockh) 1992;72:115-9. 40. Lennard L, Thomas S, Harrington CI, et al. Skin cancerin renal transplant recipientsis associatedwith increasedconcentrationsof 6-thioguaninenucleotidein red blood cells.Br J Dermatol 1985;113:723-9. 41. Streilein JW. Immunogenetic factors in skin cancer. N Engl J Med 1991;325:884-6. 42. Goldstein GD, Gollub S, Gill B. Cutaneouscomplications of heart transplantation. J Heart Lung Transplant 1986; 5:143-7. 43. Rudlinger R, Smith IW, Bunney MH, et al. Human papillomavirusinfections in a group of renal transplant recipients. Br J Dermatol 1986;115:681-92. 44. Van der Leest RJ, Zachow KR, Ostrow RS, et al. Human papillomavirusheterogeneity in 36 renal transplant recip ients. Arch Dermatol 1987;123:354-7. 45. Barr BB, Benton EC, McLaren K, et al. Human papillomavirus infection and skin cancer in renal allograft recipients. Lancet 1989;1:124-8. 46. Pfister H, GrossG, Hagedorn M. Characterization of human papillomavirus 3 in warts of renal allograft patient. J Invest Dermatol 1979;73:349-53. 47. Lutzner M, Croissant0, DucasseMF, et al. A potentially oncogenichuman papillomavirus (HPV-5) found in two renal allograft recipients.J Invest Dermatoll980;75:353-6. 48. Lutzner MA, Orth G, Dutronquay V, et al. Detection of human papillomavirus type 5 DNA in skin cancersof an immunosuppressedrenal allograft recipients.Lancet 1983; 21422-4. 49. Smith SE, Davis IC, Leshm B, et al. Absenceof human papillomavirusin squamouscell carcinomasof nongenital
Journal of the American Academy of Dermatology Volume 32, Number 3
Wittenberg,Oursler, and Peters
skin from immunccompromisedrenal transplant patients. Arch Dermatol 1993;129:1585-8. 50. Demetrick DJ, Inoue M, Lester WM, et al. Human papillomavirus type 16 associatedwith oral squamouscarcinoma in a cardiac transplant recipient. Cancer 1990; 66:1726-31. 51. Sontheimer RD, BergstresserPR, Gammas P, et al. Perturbation of epidermal Langerhans cells in immunosup
pressedhuman renal allograft recipients.Transplantation 1984;37:168-74. 52. McGregor JM, Reddi G, MacDonald D, et al. HLA-Al 1 in renal allograft recipientswith skin cancer.J Invest Dermatol 1991;98:261-2. 53. CzarneckiD, Watkins F, Leahy S, et al. Skin cancersand HLA frequenciesin renal transplant recipients.Dermatology 1992;185:9-11.
Secondary amyloidosis complicating psoriasis Gregory P. Wittenberg,
MD, Judith R. Oursler, MD, and Margot
S. Peters, MD
Rochester, Minnesota Background: Amyloidosis is a heterogeneous group of disorders characterized by extracellular deposition of fibrillar protein. Secondary amyloidosis occurs in patients with chronic infectious or inflammatory processes. Only 18 cases of secondary amyloidosis complicating
psoriasis have been reported. Objective: We characterized secondary amyloidosis complicating psoriasis. Methods: We reviewed all cases of coexisting psoriasis and amyloidosis seen at the Mayo Clinic from 1950 to 1992. The clinical characteristics were summarized, and the literature was reviewed. Results: There were 28 casesof coexistent disease, and in five of these psoriasis was the only inflammatory condition preceding the development of secondary amyloidosis. Histopathologic confirmation with Congo red staining was available in four cases, and immunohistochemical confirmation of the characteristic amyloid A subtype was performed in two.
Conhsiont Follow-up of four patients supports the view that amyloidosis associated with psoriasis is an aggressive disease that may be fatal. (J AM ACAD DERMATOL 1995;32:465-8.) Amyloidosis is defined as the extracellular deposition of amyloid, a protein composed of fine, nonbranching fibrils that aggregate into P-pleated sheets. Amyloidosis has been divided into subtypes on the basis of biochemical and clinical characteristics. Some clinical subtypes include primary amyloidosis, heredofamilial amyloidosis, local amyloidosis, senile amyloidosis, and amyloidosis associated with long-term hemodialysis. The two most common biochemical subtypes are amyloid A (AA), found in secondary and some familial forms of the disorder, and amyloid L, found in primary and multiple myeloma-associated amyloidosis. ’ From the Department dation. Accepted Reprint Mayo Copyright
of Dermatology,
for publication
Nov.
Mayo
Clinic
Foun-
7, 1994.
requests: M. S. Peters, MD, Department Clinic, 200 First St. SW, Rochester, MN @ 1995 by the American
0190-9622/95 $3.00+0
and Mayo
Academy
16/l/61867
of Dermatology, 55905.
of Dermatology,
Inc.
Secondary amyloidosis is commonly associated with a variety of chronic inflammatory disorders such as rheumatoid arthritis, ankylosing spondylitis, and osteomyelitis. It has been only rarely reported to be associated with psoriasis. AA can be isolated from secondary amyloid. This is a unique, insoluble protein with a molecular weight of approximately 8.5 kd. It is made up of 76 amino acid residues arranged in a single chain. The putative precursor to AA is an antigenically related protein, serum AA (SAA). This is a normal soluble serum component that behaves as an acute-phase reactant. SAA associates with the high-density lipoprotein-3 subclass of serum lipoproteins and is therefore sometimes referred to as apo-SAA. It is synthesized and released by hepatocytes under the control of interleukin a. The precise regulation of the conversion of SAA to the insoluble AA protein of amyloidosis is not understood. However, the induction of AA amyloidosis is thought to be a multifactorial process influenced by 465