Objectives: Transplant immunosuppression increases the risk of cutaneous squamous cell carcinoma by 65- to 200-fold. Our objective was to investigate the impact of the type of organ transplanted on the risk and presentation of cutaneous squamous cell carcinoma.
Materials and Methods: The retrospective database of the Duke University Health System was queried to identify patients who underwent an organ transplant from 1996 to 2016. Data regarding transplant outcomes, cutaneous squamous cell carcinoma, immunosuppressive regimens, and survival were recorded. We used chi-square tests, analysis of variance, and unpaired t tests to compare the incidence and presentation of cutaneous squamous cell carcinoma among organ types.
Results: Of 3652 renal, hepatic, and cardiothoracic transplant patients identified, 142 patients developed at least 1 cutaneous squamous cell carcinoma. The incidence of cutaneous squamous cell carcinoma varied by type of organ transplanted, with 46 of 1684 (2.7%) renal transplant patients developing cutaneous squamous cell carcinoma, 33 of 804 (4.1%) hepatic transplant patients, and 63 of 1164 (5.4%) cardiothoracic transplant patients over the median follow-up time of 6.5 years. Incidence in the renal transplant versus the cardiothoracic transplant group was significantly different (P < .001). The time to presentation of cutaneous squamous cell carcinoma varied significantly by group, with the renal cohort presenting at 3.8 years compared with at 2.4 years in the cardiothoracic group and 2.1 years in the hepatic group (P < .001).
Conclusions: The type of organ transplanted confers a unique risk and presentation of cutaneous squamous cell carcinoma.
Key words : Cutaneous oncology, Heart-lung transplant, Immunosuppression, Kidney transplant, Liver transplant
Since 1988, over 650 000 renal, hepatic, and cardio-thoracic (heart and/or lung) transplants have been performed in the United States.1 From 2012 to 2016, the United Network for Organ Sharing reported a 20% increase in organ transplant procedures, suggesting a trend of increasing frequency in the years to come. The immunosuppression vital to maintaining the transplanted graft is associated with an increased risk of cutaneous squamous cell carcinoma (cSCC) by 65- to 200-fold.2-5 The cSCC that develops in the transplant population has greater incidence of morbidity and mortality versus that shown in the nontransplant population.6 With few effective systemic or preventative therapies, local excision, such as Mohs surgery, and destruction, via electrodissection and curettage, are the mainstays of treatment.
There are several specific risk factors in transplant recipients that are known to increase the frequency of cSCC. Cyclosporine, previously a mainstay of transplant immunosuppression, is known to be the most carcinogenic medication used in this population. Azathioprine and belatacept also top the list of carcinogenic agents.4,7-12 Tacrolimus, a calcineurin inhibitor, and mycophenolate mofetil increase the risk of cSCC, although to a lesser extent than the previously mentioned agents.13-15 Voriconazole, an antifungal agent, commonly used in transplant recipients due to their increased risk of invasive fungal infections, is also known to confer a higher risk of cSCC.16-18 In addition to these risk factors, transplant over the age of 50 years old has been shown to further increase the cSCC hazard ratio.19-21 Subsequently, with the increasing median age of patients at transplant, there will be a larger population of posttransplant recipients with cSCC.22 Two additional intrinsic risk factors include a history of actinic keratosis and cSCC before transplant.19-21
Few studies have investigated the impact of type of organ transplanted on the risk of cSCC.23-25 So far, none have directly compared risks shown among cardiothoracic, renal, and hepatic transplant recipients at a single institution. Furthermore, all previous studies on this topic have included fewer patients, by at least 1000 participants, than our present investigation. In this study, our aim was to investigate the impact of type of organ transplanted on incidence of cSCC.
Materials and Methods
The Duke Enterprise Data Unified Content Explorer (DEDUCE) was used to identify patients who underwent a renal, hepatic, or cardiothoracic transplant from January 1, 1996 to December 31, 2016 at the Duke University Medical Center. This time period was chosen to allow for optimal data extraction via DEDUCE. The institutional review board at the Duke University Health System approved this project prior to data retrieval. Data abstraction included patient demographics, transplant dates, transplant outcomes, history of actinic keratosis and cSCC, outcome of cSCC, medication regimen, survival, death, cause of death, and last follow-up date. Due to the overwhelming likelihood of a second cSCC, patients with a history of cSCC before transplant were excluded from the study.
We used chi-square tests and unpaired t tests to compare the incidence of cSCC between organ types. An ANOVA test was used to compare the time to presentation of cSCC between organ transplant cohorts. P ≤ .05 was considered significant.
From 1996 to 2016, 3652 patients (1264 females and 2388 males) underwent a renal, hepatic, or cardio-thoracic transplant. Our results showed a total of 1684 renal transplants, 804 hepatic transplants, and 1164 cardiothoracic transplants. Of total patients, 142 patients developed at least 1 cSCC over the median follow-up of 6.5 years. The median age at transplant was 61.9 years (range, 33.9-79.0 y). Of patients who developed at least 1 cSCC, 22 were female and 120 were male (Table 1), with 140 patients of white ethnicity. The median time to cSCC was 2.5 years (range, 0.4-11.5 y) after transplant.
The incidence of cSCC varied by type of organ transplanted. In the renal transplant group, 46 of 1684 patients (2.7%) developed at least 1 cSCC compared with 33 of 804 hepatic transplant patients (4.1%) and 63 of 1164 cardiothoracic transplant recipients (5.4%) (Figure 1). We observed no differences in distribution of males versus females between organ subtypes (P = .08). However, the cardiothoracic group did have an older median age at transplant (65.0 y) than the hepatic group (59.6 y) and the renal group (55.2 y) (P < .01).
The incidence of cSCC was significantly different between the renal transplant and cardiothoracic groups (P < .001).
When we looked at type of immunosuppression, we found no statistically significant differences with regard to cyclosporine (P = .13), tacrolimus (P = .95), or mycophenolate mofetil (P = .51) treatment between the groups (Table 2). However, cardio-thoracic transplant recipients with cSCC did have significantly increased exposure to voriconazole and azathioprine (P < .01 and P < .01). Voriconazole was used in 38% (n = 24) of the cardiothoracic cSCC group compared with 6% (n = 2) in the hepatic group and 7% (n = 3) in the renal group. Similarly, azathioprine was used in 37%(n = 23) of patients in the cardiothoracic cSCC group compared with 6% (n = 2) in the hepatic group and 20% (n = 9) in the renal group. Incidence of azathioprine use did not vary between the cardiothoracic group and renal group (P = .055).
A pretransplant history of actinic keratosis was significantly more common in the renal transplant group (9 patients [20%]) than in the cardiothoracic group (3 patients [5%]; P = .015). In the hepatic group, 3 patients (9%) had pretransplant history of actinic keratosis, which was significantly greater than that shown in the renal and cardiothoracic groups (P = .18 and P = .33, respectively). Metastatic cSCC was the cause of death in 2 renal transplant patients (0.1%), 3 liver transplant patients (0.4%), and 3 cardiothoracic transplant patients (0.3%). This finding was not significantly different between the 3 groups (P = .614).
The overall time to presentation of cSCC after transplant was 2.62 years (range, 0.3-8.5 y). The time to presentation varied significantly between organ transplant group (P < .001) (Figure 2). The median time to cSCC presentation in the renal transplant cohort was 3.8 years which was significantly different from the 2.4 years in the cardiothoracic transplant cohort and 2.1 years in the hepatic transplant cohort. There was no significant difference in time to cSCC presentation when comparing the hepatic cohort to the cardiothoracic transplant cohort to each other.
Cardiothoracic transplant recipients had a higher incidence of cSCC compared with renal transplant recipients. Although not significant, the incidence of cSCC was higher in hepatic transplant recipients than in renal transplant recipients. These findings advocate for consideration of the type of organ transplanted and not simply a history of transplant when risk stratifying patients for regular derma-tologic surveillance. Specifically, cardiothoracic, then hepatic, and then renal transplant recipients should be surveilled accordingly.
This study directly compared the incidence of cSCC between the 3 largest categories of organ transplant (liver, kidney, and heart-lung). Several findings in our study add depth to discussions on risk of cSCC by organ type. The cardiothoracic group was exposed to more carcinogenic immunosup-pression, specifically azathioprine, and to the antifungal voriconazole than the other groups. The cardiothoracic group also had transplant procedures at a later age than the other groups. These findings alone independently increase the risk of cSCC. However, the renal group had a higher percentage of patients with a history of actinic keratoses, which are known to predispose to cSCC later in life.
In previously reported lung transplant recipients, voriconazole resulted in a cSCC hazard ratio of 2.39.26 However, actinic keratoses, which can recur in up to 53% of patients, may transform into cSCC in anywhere from 0.025% to 20% of patients.27,28 It is also worth noting that a large percentage of the overall cohort was composed of males, which fits with the predisposition for cSCC in that group. Age over 50 years at transplant, another risk factor for cSCC, was common in every organ transplant group. However, when taken together, the cardiothoracic group was at higher risk for cSCC due to medication exposure, but the renal group was also at an increased risk due to their personal history of actinic keratoses. The overlapping risk factors between organ transplant groups underscore the challenges in predicting which patients will develop cSCC.
We observed significantly different presentations of cSCC after transplant among the transplant groups. Despite having a lower risk of cSCC, the hepatic group had the earliest time to presentation of cSCC after transplant. Cardiothoracic transplant patients, with the highest incidence of cSCC, were diagnosed at a median time slightly later than the hepatic group but over 1 year before the renal group. These particular findings can be used to inform referral to and dermatologic follow-up periods.
The major limitation of our study was its retrospective nature. Subsequently, we relied on information available in electronic medical records and could not randomize or standardize patients to one immunosuppression regimen. Recording the cumulative dosage, rather than exposure, to immunosuppressive agents could yield more practical information; however, this information is not readily available retrospectively. Information regarding Fitzpatrick skin phototype information and family history of skin cancer was also not available for capture. Finally, with varying degrees of confounding variables affecting each transplant group, it is not feasible to match all cohorts for demographics, such as history of actinic keratoses or exposure to potent immunosuppressive and carcinogenic medications. Nonetheless, our study is one of the largest examining the impact of type of organ transplanted on the incidence of cSCC.
The type of organ transplanted is associated with a unique risk of cSCC. Given the increased morbidity and mortality of cSCC in the transplant population, referral for dermatologic surveillance and early excision should be prioritized in cardiothoracic, hepatic, and renal transplant recipients, respectively. Cardiothoracic transplant patients were found to have the highest risk for cSCC, whereas hepatic transplant recipients presented earlier, in one of the largest single center studies investigating posttransplant cSCC.
DOI : 10.6002/ect.2018.0238
From the 1Duke University School of Medicine and the 2Duke Cancer Institute and
Department of Surgery, Duke University Health System, Durham, North Carolina,
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare. This work was presented as a Late-Breaking Abstract in the Cutaneous Oncology Session of the 2018 meeting of the American Academy of Dermatology.
Corresponding author: Paul J. Mosca, DUMC 3966, 10 Bryan Searle Drive, 466G Seeley G Mudd, Durham, NC 27710, USA
Phone: +1 9 19 684 4064
Figure 1. Percentage of Patients Developing Cutaneous Squamous Cell Carcinoma in Each Organ Transplant Group
Figure 2. Median Time to Presentation of Cutaneous Squamous Cell Carcinoma by Organ Type
Table 1. Patient Demographics by Organ Transplant Group
Table 2. Immunosuppressive Medication Exposure by Organ Transplant Group