Pretransplant Malignancy as a Risk Factor for Posttransplant Malignancy After Heart Transplantation

Abstract

Background 

Posttransplant malignancy is a major cause of death in orthotopic heart transplant recipients. Our study objective is to identify whether pretransplant malignancy increases the risk for posttransplant malignancy.

Methods 

Using the Organ Procurement and Transplantation Network/United Network for Organ Sharing database, adult (≥18 years old) recipients of a primary orthotopic heart transplant performed between 2000 and 2012 were included. The cohort was stratified according to the presence and type of pretransplant malignancy. Outcomes were posttransplant overall, skin, solid malignancies, and posttransplant lymphoproliferative disorder. Incidence of and time to first posttransplant malignancy, accounting for death as a competing event, were calculated. Fine and Gray competing risks regression was performed to examine risk factors associated with posttransplant malignancy.

Results 

Of 23,171 recipients, overall posttransplant malignancy was diagnosed in 2673 recipients (11.5%).The median time to overall, skin, solid posttransplant malignancies, and posttransplant lymphoproliferative disorder were 3.2 (1.6-5.7), 3.2 (1.7-5.6), 3.5 (1.7-5.9), and 2.5 years(0.8-5.2), respectively. Pretransplant malignancy increased the risk of posttransplant overall (subhazard ratio [SHR], 1.51; 95% confidence interval [95% CI], 1.27-1.81), skin (SHR, 1.55, 95% CI, 1.23-1.93), and solid organ malignancies (SHR, 1.54; 95% CI, 1.13-2.11). Pretransplant skin malignancy increased the risk of posttransplant skin malignancy (SHR, 2.79; 95% CI, 1.82-4.28). Pretransplant solid malignancy also increased the risk of posttransplant skin malignancy (SHR, 1.55; 95% CI, 1.07-2.25) but not of posttransplant solid tumors (SHR, 1.23; 95% CI, 0.69-2.19). Older, male, and white were also risk factors for posttransplant malignancy.

Conclusions 

A history of any pretransplant malignancy was associated with increased risks of skin and solid malignancies after transplantation. The specific type of posttransplant malignancy risk differed according to the type of pretransplant malignancy.

In Brief

Registry data from 23 171 recipients of primary orthotopic heart transplant with median follow-up time of 4 years, demonstrates that a history of any pretransplant malignancy is associated with increased risk of skin and solid malignancies after transplantation.

Posttransplant malignancy accounts for more than 20% of deaths after heart transplantation.^1,2 Heart transplant recipients have a higher incidence of posttransplant malignancies compared to other solid organ transplants.³ Identification of risk factors may prompt earlier detection of posttransplant malignancies. Among heart transplant recipients, the incidence of posttransplant malignancy varies depending on geographic location and length of follow-up.^4-9 Hsu et al⁴ reported a 5.8 % incidence of posttransplant cancer over 5 years of follow-up in Chinese heart transplant recipients, whereas a European study reported a higher incidence of posttransplant cancer of 30.8% with mean follow-up of 9.2 years.⁷ Several single center and registry studies have reported an increased risk of posttransplant cancer observed among heart recipients compared with age-matched controls from the general population.^8,10-14 Although several studies have documented the incidence and risk factors of malignancies after transplantation, few have defined pretransplant malignancy as a risk factor for posttransplant malignancy in heart transplant recipients.^15-20 The objective of our study is to identify the association of pretransplant malignancy and other risk factors on developing posttransplant malignancies in heart transplant recipients.

MATERIALS AND METHODS

Study Population

Adult (≥18 years old) recipients of a primary orthotopic heart transplant (OHT) performed in the United States between the year 2000 and 2012 were selected from the Organ Procurement and Transplantation Network /United Network for Organ Sharing (OPTN/UNOS) database (data as of March 15, 2013). We excluded patients undergoing retransplantation or multiorgan transplantation. The cohort was stratified according to the presence of pretransplant malignancy and then subgrouped into skin, solid organ, hematologic (leukemia), multiple, and unknown malignancy groups. The primary outcome was the occurrence of first posttransplant malignancy after OHT. Secondary outcomes were the risk for 3 subtypes of posttransplant malignancy: skin malignancy, solid malignancy, and posttransplant lymphoproliferative disorder (PTLD).

Statistical Analysis

Baseline characteristics and differences in frequencies among recipients who did not have a history of pretransplant malignancy and those who did were computed. These characteristics were described using proportions for categorical variables and means with associated standard deviations for continuous variables. An omnibus χ² test was performed for comparison of categorical variables and 2-sample t test was performed for continuous variables.

The median time to first posttransplant malignancy and the adjusted cumulative incidence of posttransplant malignancy, accounting for death as a competing event, was calculated. Multivariate analysis using competing risks regression was performed to examine the association between pretransplant malignancies and types of posttransplant malignancies. The primary outcome of interest was time to the first report of posttransplant cancer. Because patients who died were no longer able to experience the outcome of interest, death event was treated as a competing event. The first time to cancer was modeled with Fine and Gray competing risk methods.²¹ All patients were censored at the end of the study period on March 15, 2013. Subgroup analyses were then performed to evaluate the risk of specific subtypes of pretransplant malignancies on posttransplant malignancy. Risk factors included in the multivariate model were age, sex, race, body mass index (BMI), hypertension, diabetes, renal insufficiency, HLA-DR mismatch, human leukocyte antigen mismatch, antibody induction therapy, tacrolimus use at hospital discharge, serostatus of cytomegalovirus, Epstein-Barr virus (EBV), hepatitis B virus and hepatitis C virus, and etiology of heart failure. Results of the regression analysis were expressed in subhazard ratio (SHR) and confidence intervals. All reported P values were 2-tailed, and a P value at or less than 0.05 was considered to be statistically significant. All analyses were performed using Stata Statistical Software, Release 13.1 (StataCorp LP, College Station, TX).

RESULTS

Of 23,171 OHT recipients, 1306 (5.6%) had a history of pretransplant malignancy (Table 1). There were a large number of unspecified malignancy types (322). Of the known reported malignancies, the 3 most common pretransplant malignancies were breast cancer, skin cancer, and leukemia. Recipients with pretransplant malignancy were older in age with a higher proportion of white race and females (Table 2). These patients had higher serum creatinine levels, but lower BMIs and hypertension prevalence. Antibody induction therapy and tacrolimus for maintenance immunosuppression were prescribed more often in the pretransplant malignancy group than the no-pretransplant malignancy group. Restrictive cardiomyopathy as the cause of heart failure was more common in the pretransplant malignancy group, and ischemic cardiomyopathy was more common in the group without a pretransplant malignancy. Recipients were grouped into pretransplant malignancy categories, namely, skin (n = 165) (12.6%), solid organ (n = 617) (47.2%), hematologic (n = 160) (12.3%), multiple (n = 42) (3.2%), and unknown malignancy group (n = 322) (24.7%).

TABLE 1: Type of pretransplant malignancies

TABLE 2: Characteristics of heart transplant recipients stratified by pretransplant malignancy

At least 1 posttransplant malignancy was diagnosed in 2673 recipients (11.5%) during the study period. Skin cancer was by far the commonest posttransplant malignancy, followed by lung cancer, prostate cancer, and PTLD (Table 3). The follow-up time after transplantation was 93,105 person-years, and the median follow-up time of 4.3 years (Q₁-Q₃ = 2.0-7.8). There were 30 categories of posttransplant malignancies reported in the UNOS database. These malignancies were classified into 3 groups: skin malignancy (n = 1337) (50.0%), solid malignancy (n = 1070) (40.0%), and PTLD (n = 238) (8.9%). These 3 categories encompassed virtually all of the posttransplant malignancies, except for 28 in which the type of posttransplant malignancy was unknown (1.1%). The median time from transplant to diagnosis of overall posttransplant malignancy was 3.2 years (Quartile₁-Quartile₃ = 1.6-5.7, range = 0-11.9). For the patients with posttransplant skin malignancy, the median time to diagnosis was 3.2 years (Q₁-Q₃ = 1.7-5.6, range = 0-11.9). For the patients with posttransplant solid malignancy, the median time to diagnosis was 3.5 years (Q₁-Q₃ = 1.7-5.9, range = 0-11.7), and for the patients with PTLD, the median time to diagnosis was 2.5 years (Q₁-Q₃ = 0.8-5.2, range = 0-11.3).

TABLE 3: Type of posttransplant malignancies

A history of any pretransplant malignancy was associated with increased risk of overall posttransplant malignancy (SHR = 1.51, P < 0.01), skin (SHR = 1.55, P < 0.01), and solid organ malignancies (SHR = 1.54, P < 0.01) on multivariate analysis. However, it was not significantly associated with PTLD (SHR = 0.80, P = 0.66) (Table 4). All categories of pretransplant malignancy, except pretransplant hematologic category, were associated with an increased risk of overall posttransplant malignancy on both univariate and multivariate analysis (Figure 1A and B). Pretransplant skin malignancy was a strong predictor for posttransplant skin malignancies, which led to an increased risk for overall malignancies. These patients did not have an increased risk for solid tumors. Similarly, pretransplant solid malignancy was a predictor for posttransplant overall (SHR = 1.38, P = 0.04) and skin malignancies (SHR = 1.55, P = 0.02) but was not a predictor for posttransplant solid malignancy (SHR = 1.23, P = 0.49) or PTLD (SHR = 0.60, P = 0.61). Pretransplant hematologic malignancy was not associated with any type of posttransplant malignancy.

TABLE 4: Risk factors and adjusted competing risk regression model for posttransplant malignancies

Figure 1: A, Pretransplant malignancies and unadjusted competing risks regression for posttransplant malignancies. B, Pretransplant malignancies and adjusted competing risks regression for posttransplant malignancies.

Other recipient factors including older age, male sex, and white race were also associated with an increased risk of posttransplant overall, skin, and solid organ malignancies; but only male sex was associated with an increased risk of PTLD (Table 4). In multivariate analysis, increasing age categories were associated with an increased risk of overall, skin, and solid organ malignancies.

We also investigated other potential risk factors for posttransplant malignancy. A BMI over 30 was associated with a decreased risk of posttransplant overall and skin malignancy (SHR = 0.80 and 0.76, P < 0.01 for both). Diabetes was associated with a decreased risk of posttransplant overall (SHR = 0.81 P < 0.01), skin (SHR =0.80 P < 0.01), and solid organ malignancies (SHR = 0.79 P = 0.03). Similarly, hypertension was associated with a decreased risk of posttransplant skin malignancy (SHR = 0.81, P < 0.01). The use of antibody induction therapy was associated with an increased risk of posttransplant overall (SHR = 1.11, P = 0.02) and skin malignancies (SHR = 1.16, P = 0.01), and tacrolimus use as a maintenance therapy was associated with an increased risk of PTLD (SHR = 1.48, P = 0.05). Pretransplant EBV seropositivity in recipients was associated with an increased risk of posttransplant overall (SHR =1.18, P < 0.01) and skin malignancies (SHR = 1.28, P < 0.01), but with a decreased risk of PTLD (SHR = 0.61, P = 0.01). Hepatitis C virus seropositivity and ischemic cardiomyopathy were associated with a decreased risk of overall posttransplant malignancy (SHR = 057, P = 0.01 and SHR = 0.81, P = 0.05). Figure 2 shows the adjusted cumulative incidence of posttransplant malignancy in transplant recipients with and without a history of pretransplant malignancy.

Figure 2: Adjusted cumulative incidence of posttransplant malignancy.

DISCUSSION

With a median follow-up time of 4 years, we found that 11.6% of heart recipients developed at least 1 posttransplant malignancy. Of these, skin malignancy was the most common, accounting for 50% of all posttransplant malignancies. Pretransplant malignancy was a predictor for posttransplant overall, skin, and solid organ malignancies, but the risk for specific type of posttransplant malignancy differs according to the type of pretransplant malignancy.

The strength of our study is that it is a large-scale study that is able to define the impact of a history of pretransplant malignancy on the occurrence of posttransplant malignancy in heart transplantation. We found that pretransplant malignancy is a strong risk factor for posttransplant malignancy, in particular posttransplant skin and solid organ malignancies. Whereas our data are applicable only to heart transplant recipients, our results parallel that reported in previous studies among other organ transplant recipients that have reported that a history of a pretransplant malignancy was associated with a 2-fold increase in the risk of posttransplant cancers.^17,18 Because of the limitations in the reporting to the OPTN/UNOS database, we were not able to distinguish between recurrent versus de novo malignancies after transplantation. However, a recent article using the Cardiac Transplant Research Database with 6211 primary heart transplants showed that posttransplant recurrences of pretransplant malignancies were not common.¹⁴

Previous studies in orthotopic heart transplantation have reported that pretransplant malignancy is a risk factor for posttransplant malignancy²²; however, there is no previous comparative data on the effect of specific types of pretransplant malignancy on posttransplant malignancy. As would be expected, we found that patients with a pretransplant history of skin cancer were at increased risk for posttransplant skin malignancies. Whereas all transplant recipients are generally at a high risk for skin cancer irrespective of their pretransplant history of malignancy, patients with pretransplant skin and solid malignancy had a higher risk of skin cancer after transplant than those without a pretransplant history of such a cancer. The reasons for this are unclear and may be related to an increased propensity toward further malignancies in those who have already had previous cancer or potentially due to a history of immunosuppressive therapy related to cancer treatment.

Although each category of pretransplant malignancy, except for hematologic malignancy, was associated with a higher risk of posttransplant malignancies overall, this effect was mainly driven by posttransplant skin cancers. This was particularly true in patients with a pretransplant history of both skin malignancies and solid tumors. It is reassuring to know that patients with a pretransplant history of solid tumors did not have an increased risk for solid tumors posttransplant, indicating that the pretransplant assessment of cancer remission adopted by transplant centers is generally accurate.

The EBV seropositivity at the time of transplantation (suggestive of prior EBV infection) was associated with a decreased risk of acquiring PTLD. This finding is similar to previous studies which suggest that individuals who have EBV infection after solid organ transplantation are at an increased risk of developing PTLD.^23,24 The UNOS database lacked the data on the details and type of PTLD, and therefore it is uncertain whether this is a recurrent or de novo lymphoma.

Some recent studies that have provided insight into other factors that impact outcomes of thoracic organ recipients with pretransplant malignancies are worth mentioning. A single center study from Sweden reported that mortality and recurrence of pretransplant malignancy decrease with the pretransplant cancer-free time in lung and heart recipients, with the highest risk of death and recurrence in recipients with a cancer-free pretransplant interval of less than 1 year.¹⁹ A larger study using the UNOS database suggested that a history of pretransplant malignancy is mostly not associated with decreased survival for lung and heart transplant recipients, except for increased mortality rate among heart recipients with a previous history of hematologic malignancy.²⁰

We acknowledge certain limitations to our study. First, we lacked information on pretransplant cancer staging, treatment, and timing of the cancer in relation to transplantation. Therefore, we were only able to make broad inferences about the risks of posttransplant malignancies in patients with a pretransplant cancer history, but could not identify specific high-risk subgroups among those with pretransplant malignancies. Second, in the absence of data on drug dosing and levels, we are unable to make any firm conclusions about the role of specific immunosuppressive agents on the development of posttransplant cancers. Lastly, this study is subject to some inherent limitations of registry data, where the possibility of under-reporting of posttransplant malignancies cannot be quantified. We also could not ascertain from our data whether the increased rate of posttransplant malignancies observed in those with a pretransplant history of cancer is related to more cautious screening practices among these patients.

In conclusion, over a median follow-up time of 4 years, the incidence of posttransplant malignancy among OHT recipients in the United States was 11.5%. A history of any pretransplant malignancy was associated with an increased risk of skin and solid malignancies after transplant, but the specific type of posttransplant malignancy risk differed according to the type of pretransplant malignancy. Whereas recipients with previous skin and solid malignancies had a higher risk of posttransplant skin cancer, they were not at an increased risk for solid organ tumors or PTLD. Recipients with previous hematologic malignancies were not at higher risk for any kind of malignancy. Although transplant recipients are known to be at an elevated risk for posttransplant malignancy, our study provides valuable information on the risk of specific tumor types of posttransplant malignancy. Because screening for skin cancer in transplant recipients is relatively easy, a good strategy for regular screening of these patients after heart transplant is recommended. However, the role of more intensive follow-up for posttransplant solid cancer is still unclear. Further studies are needed to ascertain the requirement of additional screening regimens for detecting posttransplant malignancies and to assess the mechanisms behind the increased propensity for posttransplant malignancies in those with a pretransplant history of cancer.

ACKNOWLEDGMENT

This study was conducted based on OPTN data as of March 15, 2013.