Fax: (717) 531-4185
Increased incidence of melanoma in renal transplantation recipients†
Article first published online: 26 SEP 2005
Copyright © 2005 American Cancer Society
Volume 104, Issue 9, pages 1962–1967, 1 November 2005
How to Cite
Hollenbeak, C. S., Todd, M. M., Billingsley, E. M., Harper, G., Dyer, A.-M. and Lengerich, E. J. (2005), Increased incidence of melanoma in renal transplantation recipients. Cancer, 104: 1962–1967. doi: 10.1002/cncr.21404
The data reported here have been supplied by the United States Renal Data System. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy or interpretation of the U.S. government.
- Issue published online: 17 OCT 2005
- Article first published online: 26 SEP 2005
- Manuscript Accepted: 31 MAY 2005
- Manuscript Revised: 16 MAY 2005
- Manuscript Received: 9 FEB 2005
- kidney transplantation;
- risk factors;
- skin neoplasms
It is well established that the incidence of nonmelanoma skin carcinoma is increased in renal transplantation recipients. However, existing studies are not in agreement over whether patients who undergo transplantation have an increased risk of melanoma. The objective of this study was to estimate the risk of melanoma among immunosuppressed renal transplantation recipients and to determine whether that risk is associated with patient and transplantation characteristics.
The authors studied 89,786 patients who underwent renal transplantation between 1988 and 1998 using the United States Renal Data System. Age standardized (to the United States 2000 population) incidence rates for melanoma were computed as diagnoses per 100,000 population and were compared with rates from the Surveillance, Epidemiology, and End Results (SEER) data. Incidence rates also were stratified to examine differences by age and gender.
Of the 89,786 patients who underwent transplantation, 246 patients developed melanoma. The age-adjusted incidence rate of melanoma among renal transplantation recipients was 55.9 diagnoses per 100,000 population. This represented an increase in age-adjusted, standardized risk that was 3.6 times greater than the SEER population. Stratified analysis suggested that the risk of melanoma accelerated in male transplantation recipients as age increased, but the risk leveled off with age among female transplantation recipients. Finally, there was a trend for patients who experienced at least 1 acute rejection episode to develop melanoma (odds ratio = 1.34; P = 0.059).
Renal transplantation recipients were nearly 3.6 times more likely to develop melanoma than the general population. Physicians who care for renal transplantation recipients should be vigilant in screening for melanoma. Cancer 2005. © 2005 American Cancer Society.
The incidence of malignant melanoma, which is the most deadly form of skin carcinoma, is increasing at an alarming rate.1 The National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) data base suggests that there was a 619% increase in the annual diagnoses of malignant melanoma from 1950 to 2000.2, 3 It was estimated that 55,000 Americans would be diagnosed with melanoma in 2004 and that 7900 would die from the disease.4 Although it has been established that ultraviolet exposure is a major carcinogen in melanoma, other factors, including genetics and immunity, also are involved. An increased risk of melanoma has been associated with the presence of atypical nevi, an increased number of nevi, history of blistering sunburns, immunosuppression, and a personal or family history of malignant melanoma.5–8 Phenotypic characteristics of blue/green eyes, blonde or red hair, fair complexion, sun sensitivity, and an inability to tan also increase an individual's susceptibility.5 Immune surveillance appears to be crucial in protecting the body from melanoma, and work is underway to develop an effective vaccine for melanoma.9, 10
If an intact immune system serves a protective role against melanoma, then it would be expected that immunosuppressed patients would be at increased risk for melanoma. Although several studies have shown that the risk of nonmelanoma skin carcinoma is increased in transplantation patients who receive long-term immunosuppression thereapy,11–13 existing studies are not in agreement over whether a similar increase in risk applies to melanoma.
The largest study of the risk of melanoma in solid organ transplantation recipients was performed by Lindelof et al.,14 who followed 5356 patients for over 24 years. Six of those patients developed melanoma, and the authors concluded that there was no increase in risk. Jensen et al.11 studied 2561 kidney and heart transplantation recipients over a 30-year period, during which 12 patients developed melanoma. Those authors reported a 3.4-fold increase in melanoma incidence for transplantation recipients. Jain et al.15 followed 1000 consecutive patients who underwent liver transplantation for up to 8 years, and 2 of those patients developed melanoma. The authors concluded that melanoma rates were not increased compared with the general population. Kasiske et al.16 recently used data from the United States Renal Data System (USRDS) to estimate the prevalence of melanoma and a number of other malignancies in renal transplantation recipients. They compared rates from those data with several reference populations, including the general United States population and patients with end-stage renal disease (ESRD) who were on the transplantation waiting list, and reported a five-fold increase in melanoma incidence among renal transplantation recipients.
Compared with nonmelanoma skin carcinoma, the low incidence of melanoma has made it difficult to study transplantation recipients prospectively. The objective of the current study was to estimate the incidence of melanoma among renal transplantation recipients using data from a large population of patients who underwent renal transplantation and to determine whether there were independent factors that increased the risk of melanoma in this population compared with the reported incidence in the general population.
MATERIALS AND METHODS
The USRDS, as the largest registry of renal transplantation patients, combines the United Networks for Organ Sharing transplantation registry data with payment data from the Centers for Medicare and Medicaid Services, which is the primary payer for the majority of patients with ESRD.17, 18 The mission of USRDS is to collect, analyze, and distribute information about ESRD in the United States. We used USRDS data to estimate the incidence of melanoma among patients who underwent renal transplantation between 1988 and 1998. Patients in the USRDS data set with any office visit, surgical procedure, laboratory study, or pathology report that had an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code for invasive melanoma (ICD-9-CM codes 172.0–172.9) were identified as potential melanoma patients. Melanoma in situ (ICD-9-CM codes 232.0–232.9) was not included in this definition. Using the method proposed by Kasiske et al.,16 we counted a patient as having melanoma only if there was at least 1 inpatient diagnosis code for melanoma or 2 outpatient diagnosis codes no more than 1 year apart. The incidence of melanoma among transplantation recipients in the USRDS data set was compared with the reported incidence among the general United States population with age-standardized incidence rates obtained from the Surveillance, Epidemiology, and End Results (SEER) data of the National Cancer Institute.19, 20
Categorical characteristics of patients, such as gender and race, were compared using chi-square tests. Continuous characteristics, such as age, body mass index, and cold ischemia time, were compared using Student t tests.
Incidence rates (per 100,000 population) were standardized by the direct method to the year 2000 United States population. Age-specific and gender-specific rates also were calculated. Note that we did not adjust incidence rates for race or ethnicity, but we did control for those variables in a multivariate analyses. Ninety-five percent confidence intervals (95% CIs) were estimated from standard errors (se), which were computed using the following approximation: se = r/√n, where r is the incidence rate, and n is the number of patients undergoing renal transplantation. Incidence rates for the USRDS population were adjusted by the average length of follow-up before computing risk ratios to compare risk relative to the SEER population.
Finally, multiple logistic regression was used to quantify specific risks for melanoma among transplant recipients. Risks included human leukocyte antigen (HLA)-A, HLA-B, and HLA-D-related (HLA-DR) locus mismatches; whether the organ was received from a deceased donor; and whether the patient ever experienced an episode of acute rejection, which may indicate the need for increased immunosuppression and, thus, may increase the risk of melanoma. We also controlled for age, gender, race, and years since transplantation. All analyses were performed using SAS statistical software (version 8.2; SAS Inc., Cary, NC).
Between 1988 and 1998, 89,786 patients in the USRDS data base underwent their first renal transplantation. Patients were followed for up to 12 years, with an average follow-up of 3.9 years. Of the 89,786 renal transplantation recipients, 246 patients (0.275%) developed melanoma. Notice that the number of melanoma diagnoses dropped generally from 1993 to 1998 (Table 1). On average, patients with melanoma were older than patients who did not develop melanoma, as expected (52.3 yrs vs. 42.4 yrs; P = 0.0001) (Table 2), and they also were more likely to be male (77.6% vs. 60.2%; P = 0.0001), to be white (95.1% vs. 73.11%; P = 0.0001), and to have experienced graft failure (52.0% vs. 40.2%; P = 0.0002).
|Transplantation yr||Total no. of transplantations||No. of melanoma diagnoses||Percenta|
|Variable||No melanoma n = 89,540 percentages||Melanoma n = 246 percentages||P value|
|Age at transplantation (yrs)||42.4||52.3||0.0001|
|Serum creatinine (mg/dL)||2.36||2.24||0.3136|
|Cold ischemia time (hrs)||22.2||28.5||0.4048|
|HLA-A, HLA-B, and HLA-DR mismatches||3.6||3.5||0.3258|
|Died within 1 yr||5.2||0.8||0.0020|
|Died within 2 yrs||7.7||4.9||0.0972|
|Died within 5 yrs||15.0||17.5||0.2833|
Renal transplantation recipients were more likely to be male (60.2% vs. 48.9%) than the general population (Table 3). They also were more likely to be black (21.9% vs. 10.4%) and usually were older than the general population.
|Variable||USRDS n = 89,540 percentages||SEER n = 534,518,012 percentages|
|≥ 80 yrs||0.0||2.4|
Across genders, renal transplantation recipients had 3.6 (95% CI: 3.1–4.1) times the reported rate of melanoma in the general population (Fig. 1). Male renal transplant recipients were 3.8 (95% CI: 3.2–4.4) times more likely than men reported in the general population to develop melanoma (P = 0.0001). Female renal transplant recipients were 1.9 times (95% CI, 1.4–2.7) more likely to develop melanoma than women reported in the general population. It is noteworthy that, among African Americans, in whom the prevalence of melanoma is very low, the annualized incidence of melanoma in renal transplantation recipients was 17.2 times greater than the reported rate for African Americans in the general population (13.32 per 100,000 population vs. 0.776 per 100,000 population; P = 0.0001).
Age is recognized widely as an important determinant of the risk of melanoma. For men, not only did the risk of melanoma increase with age groups, but risk increased at an increasing rate (Fig. 2). This was true for both male transplantation recipients and the general population, but not for females. The risk of melanoma leveled off for women around age 45–54 years.
The cumulative rates of melanoma increased with year since transplantation for each transplantation cohort (Fig. 3). Rates appeared higher for more recent years.
Logistic regression analysis was used to quantify the risk of melanoma in a multivariate context. Table 4 shows that each additional year of age at transplantation increased the odds of melanoma by 5% (odds ratio [OR] = 1.05; P = 0.0001) among renal transplantation recipients. Women had less than half the odds for melanoma than men (OR = 0.46; P = 0.0001). Black renal transplantation recipients had 7 times lower odds of melanoma than recipients of other races (OR = 0.14; P = 0.0001). The number of HLA-A, HLA-B, or HLA-DR mismatches did not significantly affect the risk of melanoma. Patients who received organs from deceased donors had 9% greater odds of melanoma, but this effect was not significant (OR = 1.09; P = 0.595). Finally, patients who experienced at least 1 episode of acute rejection had a 34% greater odds of developing melanoma (OR = 1.34; P = 0.059), but this difference did not cross the threshold of statistical significance.
|Variable||Coefficient||OR||95% CI||P value|
|Female gender||−0.783||0.46||0.34–0.62||< 0.0001|
|Black race||−1.944||0.14||0.07–0.28||< .00001|
|0 or 1 HLA-A, HLA-B, or HLA-DR mismatches||0.218||1.24||0.85–1.82||0.258|
|5 or 6 HLA-A, HLA-B, or HLA-DR mismatches||0.017||1.02||0.76–1.36||0.9065|
Prior studies have not agreed on the correlation between renal transplantation and the risk of melanoma.11, 14, 15 Our current results using data from the USRDS suggest that renal transplantation recipients are approximately 3.6 times more likely to develop melanoma than the general population. Furthermore, the risk of melanoma appears to increase with age for men, but not for women. Our data also suggest that there are increasing trends over time. For example, in Figure 2, the number of melanoma diagnoses dropped steadily from 1993 to 1998, suggesting that, although there are some melanomas that develop immediately after transplantation, melanoma may not be evident until 4 or 5 years after transplantation. The duration of immunosuppression therapy may explain at least part of this trend.
In some sense, our results are not surprising. It is well established that many malignancies, including nonmelanoma skin carcinoma, are increased in immunosuppressed transplantation patients.11, 12, 21 In addition, it is known that patients with more melanocytic nevi are at greater risk for melanoma,6, 22 and others have demonstrated an increase in the total number of melanocytic nevi in patients with renal allografts compared with controls.23, 24
The increased rate we estimate is very similar to the 3.4-fold increase suggested by Jensen et al.11 Because melanoma is relatively rare, large patient numbers are needed to obtain a precise estimate, which makes the USRDS a valuable resource. Kasiske et al.,16 as discussed above, reported a five-fold increase in melanoma incidence among renal transplantation recipients, which is somewhat higher than our estimate. The small difference between their estimate and ours may be explained by the fact that Kasiske et al. used a different data source for their comparison population.
We recognize several limitations of our study. One advantage of the USRDS data base in studying this question is that it provides a much larger sample size than is available in prospective series. However, the presence of melanoma had to be inferred from billing records, and it is likely that the diagnosis is underreported. If patients with melanoma were not assigned the correct ICD-9-CM code, or if the treatment of their melanoma was covered by an insurance carrier other than Medicare, or if the melanoma was never detected and treated, then a patient would have been misclassified. The effect of this misclassification would be to underestimate true rates and to underestimate the risk for transplantation recipients. However, some studies have suggested that Medicare billing data reasonably identifies patients with melanoma as long as both Part A and Part B information is used.25
Another limitation is that there are some studies suggesting that melanoma is underreported in the SEER data base.1, 26 Indeed, it must be recognized that, because of potential underreporting, our comparisons were with the reported incidence of melanoma in the general population, not the actual rate. To the extent that melanoma is underreported in the SEER data set, our comparison would overstate the additional risk in the renal transplantation population. Finally, there is a lack of detail in the USRDS data set on the development of nevi after transplantation. An increase in melanocytic nevi is an important marker for the risk of melanoma,6–8 and there have been reports of a high prevalence of melanocytic nevi in renal transplantation recipients.23, 24, 27 Because there was a lack of data, we were unable to evaluate or control for this particular risk factor.
In conclusion, patients who undergo kidney transplantation and who receive long-term immunosuppression have a 3.6-fold increase in the incidence of melanoma compared with the general population. However, as discussed above, their true risk may be even greater if there are coding biases in the data. Because these patients have an increased risk of melanoma, a regular, complete skin examination should be part of their routine dermatologic care, particularly among men, who have an even greater risk. Furthermore, because they carry an increased risk of melanoma, transplantation recipients should be educated about melanoma in addition to nonmelanoma skin carcinomas. Skin carcinoma surveillance in these patients should include regular, complete skin examinations by a dermatologist, and patients should be instructed on the importance of routine self-examination of their entire body.
- 2SEER cancer statistics review, 1975–2001. Bethesda: National Cancer Institute, 2004., , , et al.
- 17United States Renal Data System. Researcher's guide to the USRDS database. Bethesda: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2004.
- 19Surveillance EaERP. Program populations (1969–2000). Bethesda: National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, 2004.