Author + information
- Received July 31, 2017
- Revision received October 23, 2017
- Accepted October 26, 2017
- Published online January 1, 2018.
- Jong-Chan Youn, MD, PhDa,∗ (, )
- Josef Stehlik, MD, MPHb,
- Amber R. Wilk, PhDc,d,
- Wida Cherikh, PhDc,d,
- In-Cheol Kim, MD, PhDe,
- Gyeong-Hun Park, MD, PhDf,
- Lars H. Lund, MD, PhDg,
- Howard J. Eisen, MDh,
- Do Young Kim, MDa,
- Sun Ki Lee, MD, PhDa,
- Suk-Won Choi, MD, PhDa,
- Seongwoo Han, MD, PhDa,
- Kyu-Hyung Ryu, MD, PhDa,
- Seok-Min Kang, MD, PhDi,∗∗ ( and )
- Jon A. Kobashigawa, MDj,∗∗∗ ()
- aDivision of Cardiology, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Republic of Korea
- bDivision of Cardiovascular Medicine, University of Utah School of Medicine, Salt Lake City, Utah
- cUnited Network for Organ Sharing, Richmond, Virginia
- dISHLT Transplant Registry, Dallas, Texas
- eDivision of Cardiology, Keimyung University Dongsan Medical Center, Daegu, Republic of Korea
- fDepartment of Dermatology, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Republic of Korea
- gDepartment of Medicine, Unit of Cardiology, Karolinska Institutet, Heart and Vascular Theme, Karolinska University Hospital, Stockholm, Sweden
- hDivision of Cardiology, Drexel University College of Medicine, Hahnemann University Hospital, Philadelphia, Pennsylvania
- iDivision of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea
- jDivision of Cardiology, Cedars-Sinai Heart Institute, Los Angeles, California
- ↵∗Address for correspondence:
Dr. Jong-Chan Youn, Division of Cardiology, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Keunjaebong-gil 7, Hwaseong, Gyeonggi-do 18450, Republic of Korea.
- ↵∗∗Dr. Seok-Min Kang, Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of Korea.
- ↵∗∗∗Dr. Jon A. Kobashigawa, Division of Cardiology, Cedars-Sinai Heart Institute, 127 S. San Vicente Boulevard, Third Floor Cardiology, A3107, Los Angeles, California 90048.
Background Malignancy is a concern in cardiac transplant recipients, but the temporal trends of de novo malignancy development are unknown.
Objectives The goal of this study was to describe the temporal trends of the incidence, types, and predictors of de novo malignancy in cardiac transplant recipients.
Methods The authors analyzed the temporal trends of post-transplant incidence, types, and predictors of malignancy using 17,587 primary adult heart-only transplant recipients from the International Society for Heart and Lung Transplantation registry. The main study outcomes included the incidence of, types of, and time to de novo malignancy.
Results The risk of any de novo solid malignancy between years 1 and 5 after transplantation was 10.7%. The cumulative incidence by malignancy type was: skin cancer (7.0%), non-skin solid cancer (4.0%), and lymphoproliferative disorders (0.9%). There was no temporal difference in the time to development according to malignancy type. However, the cumulative incidence of de novo solid malignancy increased from 2000 to 2005 vs. 2006 to 2011 (10.0% vs. 12.4%; p < 0.0001). Survival in patients after de novo malignancy was markedly lower than in patients without malignancy (p < 0.0001). Older recipients and patients who underwent transplantation in the recent era had a higher risk of de novo malignancy.
Conclusions More than 10% of adult heart transplant recipients developed de novo malignancy between years 1 and 5 after transplantation, and this outcome was associated with increased mortality. The incidence of post-transplant de novo solid malignancy increased temporally, with the largest increase in skin cancer. Individualized immunosuppression strategies and enhanced cancer screening should be studied to determine whether they can reduce the adverse outcomes of post-transplantation malignancy.
De novo malignancy is an important cause of long-term morbidity and mortality in solid organ transplant recipients (SOTRs) (1,2). The incidence of de novo malignancy has been reported to be approximately 20% after 10 years of chronic immunosuppression, and other studies have also shown an overall 2- to 4-fold elevated risk of malignancy (1–6). Cardiac transplant recipients are at particularly increased risk of developing de novo malignancies, with a risk 4-fold higher than that of renal transplant recipients (6–12). However, previous studies on malignancy after heart transplantation have had limitations, such as being single-center or single-country studies without temporal trends analysis (6,10–12).
The goal of the present study, therefore, was to characterize the incidence of, types of, and time to de novo malignancy after heart transplantation and to analyze the temporal trends of characteristics for patients with de novo malignancy according to different eras (2000 to 2005 vs. 2006 to 2011) using the data from the International Society for Heart and Lung Transplantation (ISHLT) Heart and Lung Transplant Registry. In addition, the survival rates of patients after de novo malignancy were compared with those of patients without malignancy; the objective was to determine whether patients with malignancy in the recent era had better survival compared with those in the remote era. Finally, we tried to identify risk factors associated with the development of de novo malignancy after heart transplantation. Because cancer screening is mainly performed by primary care physicians in many countries, a better understanding of cancer risk in cardiac transplant recipients would help to identify opportunities to improve post-transplant outcome in these patients.
This retrospective cohort study was conducted by using data collected in the ISHLT Heart and Lung Transplant Registry. The ISHLT Registry collects data on thoracic organ transplants performed worldwide. The ISHLT Registry requires submission of core donor, recipient, and transplant procedure variables at baseline and at yearly follow-up. We did not ascertain vital status with civic registries independent of ISHLT; however, due to close regulatory monitoring of transplant programs, the vital status of patients is typically well documented. The present analysis includes only those patients in whom malignancy status after transplant was reported.
The registry provided de-identiﬁed patient-level data on all heart transplant recipients. Because no patient or center identiﬁers were included in the ISHLT dataset, our center did not require institutional review board approval, and patient consent was not required. The analysis cohort included all primary heart-only transplants in adult recipients (≥18 years of age) performed between January 2000 and December 2011. Follow-up data were available through June 2015. There were 42,343 transplants meeting these criteria. Of these, 24,756 were excluded from further analysis, resulting in 17,587 cases for analysis. Nonmutually exclusive reasons for exclusion were as follows: multi-organ and heterotopic transplants, pre-transplant history of malignancy, death or patient survival status unknown 1 year post-transplant, malignancy status not reported post-transplant, and maintenance or immunosuppression information not reported at either discharge or 1 year post-transplant. Recipient, donor, and transplant characteristics are tabulated in Online Table 1 for patients with known malignancy status at 3 or 5 years.
Rates of malignancy development
A competing risks extension of the Kaplan-Meier method was used to estimate the rates of developing each type of malignancy between years 1 and 5 after transplantation (Table 1). The only other competing event considered in this analysis was death. The analysis was based on the first event that occurred; therefore, if patients developed malignancy before death, they had a malignancy event in the analysis rather than a death event. Rates were computed separately for each type of malignancy. Some patients may have experienced >1 type of malignancy; thus, the sum of the rates for individual malignancies may be larger than the overall rate. The results are stratified according to transplant era. Of note, because we planned to examine what effect the events in the first year post-transplant had on the risk of malignancy, and because death due to de novo malignancy is unlikely in the first year post-transplant, patients who died or were diagnosed with malignancy before 1 year after transplant were excluded.
Patient survival rates were computed via the Kaplan-Meier method and compared by using the log-rank test statistic. For patients in whom malignancy was diagnosed between years 1 and 5 after transplantation, the diagnosis date was used for time zero. The median time to diagnosis for malignancies diagnosed within 5 years after transplantation was computed for each type of malignancy. For patients who were not reported to have a malignancy within the median time for that malignancy, time zero was designated as the median time to malignancy development in the corresponding group of patients who developed a malignancy. For example, for patients who developed skin malignancy within 5 years, the median time to development was 893 days; therefore, survival rates were computed starting at day 893 when assessing survival in this control group cohort.
Cox proportional hazards regression models were used to assess the relationship of various potential recipient, donor, and transplant risk factors and the development of malignancy within 5 years, conditional on survival to 1 year (deaths before 1 year were excluded) for each malignancy type.
All continuous factors were included in the models considering the use of a restricted cubic spline to allow for the most flexible fit of the functional form. When appropriate, continuous variables were modeled only as linear terms and are specified as such. The detailed list of variables considered for inclusion in the multivariate models can be found in the Online Appendix. A backward selection method was used to determine which risk factors to retain in each model. A p value <0.05 was considered significant, and a p value ≥0.05 but <0.10 was considered borderline significant. Variables forced into the model regardless of statistical significance were recipient age, diagnosis group (categorical), and transplant era.
Statistical analyses were performed by using SAS Enterprise Guide 5.1 (SAS Institute, Inc., Cary, North Carolina) and R version 0.99.486 (RStudio Team , RStudio: Integrated Development for R. RStudio, Inc., Boston, Massachusetts).
Baseline characteristics of the study population
Baseline characteristics, including recipient, donor, and transplant characteristics, of the study population according to malignancy status at 3 years (n = 14,426) and 5 years (n = 10,829) are summarized in Online Table 1. For patients with known malignancy status at 5 years, the mean recipient age was 52.2 ± 11.9 years, and 77.3% of recipients were male. The main underlying diagnoses of heart failure leading to heart transplantation were nonischemic cardiomyopathy (47.4%) and coronary artery disease (43.9%). Twenty-eight percent of patients underwent at least 1 pre-transplant mechanical circulatory support device procedure. The mean donor age was 31.1 ± 12.2 years, and 72.2% of donors were male. The mean ischemia time was 3.2 ± 1.0 h. The most common maintenance immunosuppression included calcineurin inhibitors, either tacrolimus (49.2%) or cyclosporine (46.2%), and cell cycle inhibitors, most commonly mycophenolate mofetil (MMF) (85.6%). Almost 43% of transplant recipients were hospitalized for any reason between discharge and 1 year post-transplant.
Risk of de novo malignancy within 5 years after heart transplantation
Table 1 summarizes the rates of development of de novo malignancy after heart transplantation. The incidence of any de novo solid malignancy between years 1 and 5 after transplantation throughout the entire period was 10.7%. The cumulative incidence of de novo solid malignancy was higher in the 2006 to 2011 cohort compared with the 2000 to 2005 cohort (12.4% vs. 10.0%; p < 0.0001). This increase was predominantly owing to the higher incidence of skin cancer in the more recent cohort (8.4% in 2006 to 2011 vs. 6.4% in 2000 to 2005; p < 0.0001). The cumulative incidence of non-skin solid cancer was also higher in the more recent era, but the degree of this increase was relatively small (4.5% in 2006 to 2011 vs. 4.0% in 2000 to 2005; p = 0.004). The incidence of lymphoproliferative disorders was not significantly different between the 2 cohorts (0.9% in 2006 to 2011 vs. 1.0% in 2000 to 2005; p = 0.1118). Within the de novo skin cancer group, the incidence of squamous cell carcinoma (SCC) increased from 4.0% to 5.9%, and the incidence of basal cell carcinoma from 3.1% to 3.5%, both of which were statistically significant (p < 0.05). Within the de novo non-skin solid cancer group, the most frequent malignancies were prostate cancer (1.4% in 2006 to 2011 and 1.3% in 2000 to 2005) and lung cancer (1.0% in 2006 to 2011 and 1.1% in 2000 to 2005).
Time to new diagnosis of malignancy after transplant
Given the change in the cumulative incidence of malignancy, we were interested to know whether the time from transplant to diagnosis differed between transplant eras. For any malignancy, the median time to diagnosis was similar for the 2 eras: 899 days in 2001 to 2005 and 900 days in 2006 to 2011 (p = 0.6605). We also explored time to malignancy in the 2 eras between the different malignancy subgroups and found no significant differences according to malignancy type (Figure 1).
For all types of cancer, survival rates of patients after de novo malignancy were markedly lower than those of patients without malignancy diagnosed within the median time of cancer diagnosis for the respective cancer types (Figure 2). All comparisons of patient survival rates between those who developed a malignancy between years 1 and 5 (using time at cancer diagnosis as time zero) and those who did not develop a malignancy (designating time zero as the median time to diagnosis in the former group) were statistically significant. When the patient survival rates after malignancy development were stratified according to transplant era, higher survival was seen in the more recent era, but this finding was not statistically significant for any malignancy type (Figure 3).
Risk factors for post-transplant malignancy
Three multivariable proportional hazards models were used to assess the association between various potential recipient, donor, and transplant factors and the risk of developing de novo skin malignancy, non-skin solid malignancy, and lymphoproliferative disorder after heart transplantation.
For skin malignancy, recipient age and transplant era had large effects, with older recipients and those undergoing transplantation more recently having a higher risk of de novo skin malignancy within 5 years (Table 2). Additional risk factors for the development of skin malignancy within 5 years of transplant included larger height, use of interleukin-2 receptor antagonist or muromonab-CD3 induction, hospitalization between discharge and 1 year post-transplant, human leukocyte antigen DR mismatches (1 or 2 vs. 0), donor/recipient cytomegalovirus mismatch, use of azathioprine versus MMF 1 year post-transplant, and congenital heart disease or retransplant/graft failure diagnoses versus cardiomyopathy. Similar to skin cancer, the risk factors for development of non-skin solid cancer within 5 years included age, more recent transplantation, height, and hospitalization between discharge and 1 year post-transplant. Several additional risk factors were identified that were unique, including a recipient history of smoking and the presence of drug-treated systemic hypertension. Risk factors for the development of de novo lymphoproliferative disorders included no cell cycle inhibitor use versus MMF or azathioprine versus MMF use at 1 year follow-up, overweight, negative Epstein-Barr virus serostatus, hospitalization between discharge and 1 year post-transplant, and use of antithymocyte globulin induction.
In this cohort of 17,587 adult cardiac transplant recipients from the ISHLT Registry, >10% developed de novo malignancy between years 1 and 5 after transplantation, which was in turn associated with a significantly increased risk of mortality (Central Illustration). The incidence of de novo malignancy increased in the recent era (2006 to 2011 vs. 2000 to 2005), with the largest increase seen in skin cancer. Survival rates of patients after de novo malignancy were markedly lower than those of patients without malignancy for all types of cancer. Importantly, the increased risk of mortality was sizeable even for patients diagnosed with skin cancer; this finding is in contrast to the general population, in whom survival after skin cancer is typically favorable. When the patient survival rates after de novo malignancy were stratified according to transplant era, survival rates were higher in the more recent era, but this finding was not statistically significant for any malignancy type. Multivariate analysis revealed that primarily older recipients and patients who underwent transplantation more recently had a higher risk of de novo skin cancer and non-skin solid cancer.
Skin cancers account for >40% of malignancies in organ transplant recipients, and they include SCC, basal cell carcinoma, and melanoma (13). The risk for the development of non-melanoma skin cancer is known to increase >10-fold in chronically immunosuppressed patients who undergo solid organ transplantation (14). Reduced immune surveillance, the direct carcinogenic effect of immunosuppressive agents, and proliferation of oncogenic viruses may contribute to the development of skin cancer in these patients. It is well known that the incidence of cutaneous SCC increases with the duration and degree of immunosuppression (15–18). Chronic immunosuppression may increase the incidence of cutaneous SCC and, to a lesser extent, basal cell carcinoma. In the United States, approximately 20% of heart transplant recipients will develop skin cancer within 10 years after transplantation. It is therefore plausible that more vulnerable older recipients and the more intensified immunosuppression in recent years may have precipitated the increase in the incidence of de novo skin cancer, including SCC. The risk of malignancy is of increasing concern because early survival continues to improve in heart transplant recipients, and malignancy becomes relatively more important than other causes of morbidity and mortality with increasing time post-transplant (19). In addition, as an increasing number of older patients receive heart transplants (20,21), including after mechanical circulatory support, this population may be particularly vulnerable (22).
Another possible explanation for the increased incidence of de novo solid malignancies in the recent era, especially for skin cancer, can be found in the increasing incidence of skin cancer in the general population. Because cutaneous SCC and basal cell carcinoma are not typically reported to cancer registries, the exact incidences of these malignancies are unknown. However, recent studies revealed that the incidence of non-melanoma skin cancer, including SCC and basal cell carcinoma, is increasing worldwide (23–27). Although skin cancer in the general population exhibited a high and stable survival rate in 2006 to 2011 (27), the survival rate of patients who underwent cardiac transplants and were diagnosed with de novo skin cancer was markedly lower, however, than that of cardiac transplant patients without skin cancer in our study.
Recently, Acuna et al. reported that SOTRs are at increased risk of cancer-specific death, regardless of age, sex, organ transplanted, or transplant period, which could serve to justify pursuing targeted cancer screening in these patients (28,29). However, controversy still exists about such targeted cancer screening programs, owing to the reduced life expectancy and competing causes of death in these patients (30). Although there are some established recommendations for cancer screening in SOTRs, recommendations vary among organizations and are generally limited to kidney recipients. Although there is consensus on the recommendations for skin cancer screening in the general population, and these recommendations extend to SOTRs, recommendations for screening in other malignancies are highly variable.
ISHLT guidelines recommend that cardiac transplant recipients have close skin cancer surveillance, including education on preventive measures and yearly dermatologic examinations. Recommendations regarding screening for breast, colon, and prostate cancer in the general population should also be followed in cardiac transplant recipients (31). In addition, it is recommended that chronic immunosuppression be minimized where possible, particularly in patients at high risk for malignancy. Some data suggest that higher recipient age is strongly associated with increased risk of death from infection and malignancy, whereas it is associated with reduced risk of death from acute rejection, cardiac allograft vasculopathy, and graft failure (32). In addition to reduction of chronic immunosuppression, it would seem that avoidance of immunosuppression induction in patients at high risk of de novo malignancy after transplant would be advisable. However, there have been few efforts to systematically tailor immunosuppression according to age or cause-specific morbidity and mortality risk.
Considering the increased burden of de novo malignancy in cardiac transplant recipients, additional effort needs to be directed toward formulating evidence-based cancer screening recommendations and optimized immunosuppression protocols for these patients. Relevant stakeholders, including oncologists, primary care physicians, and public health experts, as well as transplant cardiologists and immunologists, might be involved in the formulation of screening recommendations. In addition, it may be reasonable to consider the risk of de novo post-transplant malignancy in older patients when making decisions regarding candidacy for heart transplant versus left ventricular assist device as destination therapy.
There is no guideline or consensus statement available to direct cancer surveillance methods after heart transplantation. Therefore, cancer surveillance methods might have varied according to each institution’s strategies. Second, we could not verify the detailed information regarding the stage, subtypes, and treatment of individual malignancies, as well as the ethnicity of the recipients, due to limitations of the ISHLT Registry data. Finally, we also could not confirm the detailed underlying mechanisms of the increased incidence of de novo malignancy in this study. We cannot rule out that the increased incidence of cancer in the most recent cohort of heart transplant recipients resulted from better cancer screening in these patients. However, even if this scenario is the case, the implications of our study do not change, as the survival of patients with malignancy, including skin cancer alone, still remains far below the survival of those without cancer, even in the more recent patient cohort.
This large international registry study documents the risk of a wide spectrum of de novo malignancies and temporal trends in malignancy incidence after heart transplantation. More than 10% of recipients developed de novo malignancy between years 1 and 5 after transplantation, which resulted in a significantly increased risk of mortality. Based on our findings, further research is necessary to investigate the best approaches for prevention and early detection of de novo malignancy. Individualized immunosuppression and intensified cancer screening, especially for skin cancer, should be studied to determine whether these approaches can reduce the adverse outcomes of post-transplantation malignancy.
COMPETENCY IN MEDICAL KNOWLEDGE: Heart transplant recipients are at increased risk of developing de novo malignancies, and those who develop malignancy face significantly shortened survival.
TRANSLATIONAL OUTLOOK: Further studies are needed to develop individualized immunosuppression strategies for heart transplant recipients to reduce the frequency of de novo malignancies and improve survival.
The authors acknowledge the contributions of Leah B. Edwards, PhD, to the manuscript.
This research was supported by the 2015 Transplant Registry Early Career Award Grant from the International Society for Heart and Lung Transplantation and by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT and Future Planning (NRF-2015R1C1A1A02036645). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Dr. Stehlik has served as a consultant to Medtronic. Dr. Kobashigawa has received a research grant from Novartis. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- International Society for Heart and Lung Transplantation
- mycophenolate mofetil
- squamous cell carcinoma
- solid organ transplant recipient
- Received July 31, 2017.
- Revision received October 23, 2017.
- Accepted October 26, 2017.
- 2018 American College of Cardiology Foundation
- Kapoor A.
- Frohlich G.M.,
- Rufibach K.,
- Enseleit F.,
- et al.
- Rossano J.W.,
- Dipchand A.I.,
- Edwards L.B.,
- et al.
- Youn J.C.,
- Han S.,
- Ryu K.H.
- Ciarka A.,
- Edwards L.,
- Nilsson J.,
- Stehlik J.,
- Lund L.H.
- Callens J.,
- Van Eycken L.,
- Henau K.,
- Garmyn M.
- Acuna S.A.,
- Fernandes K.A.,
- Daly C.,
- et al.
- Schmid M.,
- Chun F.K.,
- Trinh Q.D.
- Acuna S.A.,
- Huang J.W.,
- Scott A.L.,
- et al.
- Wever-Pinzon O.,
- Edwards L.B.,
- Taylor D.O.,
- et al.