The all-cause mortality rates of adult recipients after liver transplantation (LT) have incrementally improved in the last 4 decades, with current recipient survival rates in the United States and elsewhere approximating 90% at 1 year and 70% at 5 years.1, 2 However, the greatest reduction in recipient mortality in the last 2 decades has occurred largely in the first year after transplantation, and survival rates beyond 1 year have been relatively stagnant during the same period.3, 4 For the recipient who survives hepatic transplantation without allograft or postoperative complications, malignancies are among the most frequent causes of morbidity and death.3, 5, 6
Malignancies account for 3 of the top 10 causes of death in developed countries, and they are a tremendous global burden.7 Multiple observational studies have shown a 2-fold increase in the rate of solid organ malignancies and a 30-fold or higher increase in the rate of lymphoproliferative malignancies in comparison with the general population.8-11 The higher incidence of de novo malignancies in the post-LT population may not be surprising because recipients receive long-term immunosuppression therapy on top of other risk factors associated with carcinogenesis.
This review briefly summarizes our current understanding of the epidemiology, risk factors, pathogenesis, and management of post-LT malignancies. In addition, we aim to identify the gaps in medical knowledge that warrant prospective studies when they are feasible.
ALD, alcoholic liver disease; CHOP, cyclophosphamide, adriamycin, oncovin, and prednisone; CRC, colorectal cancer; EBV, Epstein-Barr virus; HPV, human papillomavirus; IBD, inflammatory bowel disease; LT, liver transplantation; mTOR, mammalian target of rapamycin; NIDDK, National Institute of Diabetes and Digestive and Kidney Diseases; NMSC, nonmelanoma skin cancer; PSC, primary sclerosing cholangitis; PTLD, posttransplant lymphoproliferative disorder; SIR, standardized incidence ratio.
Posttransplant lymphoproliferative disorder (PTLD) refers to uncontrolled lymphoproliferation in an immunocompromised host after solid organ transplantation. In comparison with the general population, lymphoproliferative disorders occur with a standardized incidence ratio (SIR) of 7.77 after LT.10 The rate of PTLD is lower in LT recipients (1%-2%) versus recipients of other solid organ transplants,12 and this can likely be explained by the lower immunosuppression requirements of LT recipients. The incidence of PTLD is greatest in the first 12 to 18 months after LT.12 Additional established risk factors for PTLD are listed in Table 1.19
Table 1. Risk Factors for PTLD
Recipient age less than 18 years
The pediatric incidence rate is 3 or more times higher than the rate in adult recipients
The pathogenesis of PTLD is not fully elucidated, but Epstein-Barr virus (EBV) plays a strong etiological role. EBV is a ubiquitous herpesvirus that can stimulate B cell proliferation and transformation. In immunocompetent individuals, the vast majority of whom encounter EBV by early childhood, the primary infection occurs in the oropharynx and causes humoral and cellular immune responses. After clearance, EBV establishes a latent infection with partial genome expression in some host B cells, but a robust immune system keeps the virus dormant.20 In an immunosuppressed LT recipient, however, the function of EBV-specific cytotoxic T lymphocytes is impaired, and this weakens the surveillance capacity of T cells and enables the propagation of EBV-infected lymphocytes, which ultimately culminates in the development of lymphoma.20 For the LT recipient who has never been exposed to EBV, a primary EBV infection can be catastrophic because the recipient cannot readily generate EBV antibodies to neutralize the infection. Less commonly, although with increasing frequency, PTLD can occur in the absence of EBV.21 EBV-negative PTLD typically presents later in the posttransplant course, is monomorphic, and carries a worse prognosis. The mechanism of EBV-negative PTLD is poorly understood.21, 22
PTLD is almost always derived from recipient lymphocytes rather than donor lymphocytes. It most commonly involves a polyclonal growth pattern, although monoclonal replication can ensue via mechanisms of natural selection and heralds a worse prognosis.23, 24 Lymphoproliferation of a B cell origin constitutes approximately 85% of cases, 80% of which are associated with EBV.25 The remaining 15% of PTLD cases have a T cell origin, but only 30% of these cases are EBV-related.26
PTLD has a broad and heterogeneous clinical spectrum ranging from infectious mononucleosis with reactive polyclonal hyperplasia to systemic, high-grade monoclonal lymphoma with nodal and extranodal involvement.20, 27 In comparison with nontransplant subjects with lymphoma, patients with PTLD are more likely to have extranodal involvement, high-grade and aggressive lymphoma, and poor clinical outcomes. Negative prognostic factors are listed in Table 2.
Clinical manifestations correlate with the disease location and stage and the performance status of the patient. Often, patients may have nonspecific symptoms such as anorexia, coughing, diarrhea, headache, nausea, vomiting, and unintended weight loss. The most common sites of involvement are the lymphatic system, the gastrointestinal tract (including the hepatic allograft), and the kidneys. EBV-negative T cell disease can involve extranodal organs, including the central nervous system. Clinical deterioration can be rapid in these cases.
A timely diagnosis at an early stage is essential for preventing advanced progression and for improving outcomes. In addition to a high index of suspicion, laboratory parameters such as elevated lactate dehydrogenase levels and high/rising EBV viral loads and diagnostic imaging demonstrating characteristic irregular nodes and extranodal infiltrating masses are strongly suggestive. The potential role of routine EBV monitoring in facilitating the early diagnosis of PTLD has not been proven beneficial.35 In fact, only 50% of patients found to have increased EBV viral levels on routine screening developed overt PTLD, so its role as a screening test is limited.36 A tissue diagnosis, preferably with excisional or needle core biopsy (fine-needle aspiration generates inadequate/nondiagnostic samples), should be sought to confirm the diagnosis and identify the cell type involved. Tissue analysis allows molecular stratification, which can shed light on the prognosis and help to guide therapy. Once the diagnosis of PTLD is made, comprehensive staging with body computed tomography and ancillary testing such as echocardiography, bone marrow biopsy, and positron emission tomography scanning should be considered on a case-by-case basis.28
In a large series of 7040 solid organ recipients at the University of Michigan, the median overall survival for 78 patients with PTLD was 8.23 years (95% confidence interval = 2.28-30.0 years).15 In a 20-year experience with 170 recipients who developed PTLD (out of a total of 4000 recipients) at the University of Pittsburgh, the actuarial patient survival rates 1, 5, and 10 years after the diagnosis were 85%, 69%, and 55%, respectively.37 Although the incidence of PTLD may have been higher in the tacrolimus era in the Michigan experience,15 the overall survival rates improved in the tacrolimus era,37 and this was presumed to be related to the lower acceptable immunosuppressive levels and improved PTLD treatments. In this series, 44% of the 80 deaths were secondary to PTLD, with the second most common cause of death being infection. Because of the retrospective nature of the study, it is unclear whether infections developed as a result of PTLD treatment. The higher mortality rates associated with PTLD are related to the age of the patient, with children showing better long-term survival than adults, and to the timing of the diagnosis, with advanced disease leading to higher treatment failure and mortality rates than early disease.37 The recurrence of PTLD after remission has been described in up to 13% of patients, with the majority having an EBV-negative state before transplantation.38
The treatment of PTLD is led by hematological oncologists in conjunction with the transplant team. PTLD is a heterogeneous disease, and proposed treatment regimens are largely based on expert opinion, case series, and/or phase 2 trials.39 The first step in management is the reduction of immunosuppression.28 The risk of rejection of the hepatic allograft must be weighed against the immediate threat of PTLD to the patient's well-being and survival. In general, because the recipient's immune system is somewhat tolerogenic to the hepatic allograft and acute cellular rejection is highly responsive to therapy, a 25% to 60% dose reduction in immunosuppression is frequently sufficient to facilitate remission while protecting the patient against rejection.40, 41 Some patients tolerate the complete withdrawal of immunosuppression, but it is unclear whether the rate of acute cellular rejection (quoted at 40%40) is more frequent in these patients. Immunosuppression reduction alone can result in up to a 90% response rate within 2 to 4 weeks in patients with early polymorphic disease without high lactate dehydrogenase levels, organ dysfunction, or multiorgan involvement, but the rate is only 60% with any of these features and 0% with more than 1 of these features.28 There are limited observational data on the overall effectiveness of immunosuppression reduction alone in inducing the remission of PTLD, but monocentric experiences have approximated the rate of success of this strategy at 45% to 70%.40-42 In the pediatric population, a small study following EBV viral replication after transplantation suggested a reduced incidence of PTLD with immunosuppression dose reduction once EBV replication was noted.43
For patients with infectious mononucleosis or polyclonal lymphoproliferation with early malignancy transformation, the addition of antiviral therapy may be helpful, but there is no evidence supporting the use of antiviral therapy in those with advanced PTLD.44-46
Other treatments, including chemotherapy, rituximab, radiation therapy, and surgery, are considered when the reduction of immunosuppression alone is insufficient [or is felt to be insufficient (ie, in advanced disease)] to induce remission. Surgery and radiation therapy are generally used only in patients with localized disease (refractory to reduced immunosuppression). Radiation therapy may be particularly applicable to lymphoma in the central nervous system, which is often difficult to treat and has an overall mortality rate of 88%.47, 48
Rituximab, a monoclonal antibody to CD20, is a key agent in the arsenal of PTLD therapy. CD20 is a B lymphocyte–specific antigen frequently expressed in the majority of cases of PTLD.15 In subjects with CD20+ expression, rituximab leads to remission in 44% to 65% of cases.49, 50 After a failed response with immunosuppression reduction, many centers use rituximab as a second-line treatment for CD20+ PTLD because of its safety and comparable effectiveness to chemotherapy.51 Data on more than 500,000 patients receiving rituximab have shown a favorable safety profile with a low incidence of severe adverse events, and this has led many experts to advocate rituximab as the first-line therapy instead of chemotherapy for those with PTLD refractory to immunosuppression reduction alone.52 For patients who fail to have a complete response to rituximab, salvage chemotherapy can be considered. In a prospective study by Trappe et al.,53 the role of sequential rituximab followed by cyclophosphamide, adriamycin, oncovin, and prednisone (CHOP) therapy for adults with B cell PTLD was assessed in 70 subjects, with 90% of the patients experiencing a complete or partial response.
Various chemotherapeutic regimens have been tried in patients with PTLD, and although no individual regimen has been identified as the gold standard, single-agent regimens appear inferior to combination regimens.54 CHOP—the most widely used and best evaluated regimen—is associated with a response rate of approximately 65%.55 CHOP or comparable regimens (eg, prednisone, methotrexate, doxorubicin, cyclophosphamide, and etoposide) should be considered in patients with an inadequate response to other therapies, high-grade lymphoma, or organ compromise. Alternative multidrug chemotherapeutic agents have been tried in lieu of CHOP, but there has been no demonstration of superior efficacy or safety.54
Overall, a better understanding of the pathogenesis and risk factors of EBV-negative PTLD are needed. Improvements in screening tools to better identify early PTLD are required to maximize patients' success with treatment. Continued investigation into salvage chemotherapeutic options for the more aggressive cases is needed.
SOLID ORGAN MALIGNANCIES
The risk of de novo solid organ malignancies among LT recipients is approximately 1% per annum, with an overall incidence of 3% to 15%.5, 56 Unlike PTLD, solid organ malignancies occur more commonly after the first posttransplant year, do not have a predisposition for children, and are associated with increasing recipient age. Large registry studies10, 11 suggest that LT recipients are 2 to 3 times more likely to develop de novo malignancies than the general population. Specific patient populations within the LT community, including those with a history of alcoholic liver disease (ALD) and primary sclerosing cholangitis (PSC), may be at highest risk.56 The established risk factors for de novo solid organ malignancies are outlined in Table 3.
Table 3. Risk Factors for Solid Organ Malignancies After LT
Of all the malignancies that can occur in LT recipients, nonmelanoma skin cancer (NMSC) is by far the most common with a SIR > 30.9 The overall incidence of NMSC is 16% to 22.5%, and it can present any time after transplantation.60, 61 In general, NMSC does not affect mortality. In fact, 1 study using the United Network for Organ Sharing database found that patients with NMSC had a lower risk of dying within 5 years after LT than patients without NMSC; this was thought to be most likely a reflection of overall immunosuppression and a lower risk of allograft rejection or failure.62 The risk of skin cancer is largely based on a patient's age, skin type (types that tend to burn more than tan), and lifetime sun exposure, with some studies suggesting male sex, red hair, and brown eyes as added risk factors.56, 60, 61, 63 PSC has been associated with increased risk as well, but only in a few studies.56, 60 Immunosuppression increases the risk of skin cancer, but whether cyclosporine poses a higher risk in comparison with tacrolimus is unclear. Sirolimus, however, may reduce the risk in comparison with these other agents (see the section on immunosuppression).
It is recommended that all recipients undergo routine dermatological examinations by their health care providers with a low index of suspicion for biopsying and/or removing any concerning skin abnormalities. Notably, many case series have reported multiple NMSCs within individual patients either concurrently or sequentially, and this emphasizes the need for particular vigilance in patients with an index event of NMSC.64
De novo nonskin solid organ malignancies, while less common, are an important threat to recipient survival after LT. In a prospective, observational study using the LT database of the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) from 1990 to 2003, 95 of 798 adult LT recipients (12%) developed nonskin solid organ malignancies.56 The UK transplant database demonstrated an overall SIR of 2.2 for its LT recipients with respect to the general population, with oral cavity, colorectal/anal, and kidney cancers most commonly seen.11 In a study comparing cancer rates in the general Finnish population and 540 LT recipients, Aberg et al.9 found the risks for colorectal cancer (SIR = 3.6), kidney cancer (SIR = 8.5), and central nervous system cancer (SIR = 12.6) to be most dramatically elevated. Similarly, an Italian study found the overall risk of de novo malignancies after LT to be 3 times higher than the risk in the general population, and risks were particularly elevated for Kaposi's sarcoma (SIR = 212), esophageal cancer (SIR = 18.7), and head and neck cancers (SIR = 4.6).65 According to these studies, LT recipients appear to be at increased risk for GI malignancies (particularly esophageal and colorectal malignancies), head and neck cancers (with special concern for oropharyngeal/laryngeal cancers), lung cancer, and genitourinary cancers (including cervical, vulvar, and kidney cancers); only 1 of these studies showed an increased risk of central nervous system tumors. SIR values determined in several of these studies have been adapted in Fig. 1. The overall risk of malignancies such as Kaposi's sarcoma must be considered in context because they are almost unheard of in the general immunocompetent population.
Risk factors for nonskin solid organ malignancies after LT include age, immunosuppression, noninfectious environmental exposures (the most established of which are alcohol and smoking), infectious agents (discussed later), and underlying liver disease.56, 59
Alcohol exposure is a well-established risk factor for several cancers, including oropharyngeal, laryngeal, esophageal, and liver malignancies among others.67 The purported mechanisms of alcohol-mediated oncogenesis are poorly understood, but they may involve the carcinogenic properties of acetaldehyde and/or the inhibition of DNA methylation via the alteration of retinoid processing.68 Despite molecular and epidemiological evidence for its role in cancer development, there are no high-level data to quantify the risk of alcohol for cancer development in either the pre-LT setting or the post-LT setting. Adult recipients undergoing transplantation for ALD have been noted to have an almost 2-fold higher rate of nonskin cancer development in 10 years in comparison with adult patients undergoing transplantation for other hepatic conditions, with the exception of PSC.56 An NIDDK analysis reported a cumulative incidence of nonskin cancer at 10 years of 18% in ALD recipients versus 10% in non-PSC, non-ALD recipients.56 Patients with ALD have also been noted to have a significantly higher rate of colonic adenoma formation after LT.69
In addition to ALD, patients undergoing transplantation for PSC have a relatively high rate of de novo malignancies (in fact, the highest rate of all liver disease groups) with a cumulative incidence of nonskin solid organ cancers of 22% in 10 years.56 Because a sizeable proportion of patients with PSC also have inflammatory bowel disease (IBD), higher rates of colorectal cancer (CRC) after transplantation in comparison with nontransplant matched controls could be anticipated. LT patients with PSC and IBD had an estimated hazard ratio for the development of CRC of 3.51 (P = 0.005).56 In a case-control study, Hanouneh et al.70 found that PSC/IBD subjects who underwent LT had the same rate of CRC as PSC/IBD patients who did not undergo LT, and the 2 groups had similar survival rates after the diagnosis of CRC.70 Other studies have also suggested an increased incidence of colon cancer in this population.71, 72 One study suggested that PSC patients were at higher risk for skin and hematological malignancies,56 but this has not been verified by other investigators.
Infectious exposures are important risk factors for several cancers that are seen at higher frequencies in transplant recipients. A summary of important infectious agents and their associations with particular malignancies is presented in Table 4. One classic example is Kaposi's sarcoma, which has an associated SIR of 212 in some populations.65 Kaposi's sarcoma is a malignancy exclusively seen in the immunocompromised population, and it arises from endothelial cells. Human herpesvirus 8 infection plays an etiological role. Antiviral agents have shown some benefits,73 as has sirolimus in case reports (see the section on immunosuppression management). Human papillomavirus (HPV) has increased in the general population and has been identified on anal screening in up to 29% of LT recipients before transplantation.74 HPV is problematic in the immunosuppressed population because it increases the risk for future genitourinary, anal, oropharyngeal, and skin cancers, and it has been noted on cervical screening tests in up to 62% of female renal transplant recipients.75 The prevention of viruses with oncogenic potential is problematic because of their pervasiveness, and should an active infection occur, the general management approach is a reduction in immunosuppression. Unfortunately, vaccines have not yet been established for most infections with oncogenic potential (with an exception now for HPV). Vaccination for transplant recipients either before immunosuppression (preferable) or after transplantation should be investigated further. Active hepatitis B replication at the time of transplantation increases the risk of a posttransplant recurrence of hepatocellular carcinoma,76 but little is known about the risk of de novo hepatocellular carcinoma. Hepatitis C recurrence with advanced fibrosis has been associated with de novo hepatocellular carcinoma in at least 1 case report.77
Table 4. Summary of Infectious Agents and Their Associated Risks for Solid Organ Cancers
The theoretical risk as an association in the posttransplant setting is unclear.
Human herpesvirus 8
Cervical, vulvar, vaginal, anal, penile, and oropharyngeal cancers
Prostate cancer and breast cancer, which are among the most common malignancies in the general adult population, do not appear to be increased in frequency in LT recipients versus age-matched controls.57, 78 Although CRC is more frequent in the LT population10, 11 the literature is mixed on whether all LT recipients, in the absence of a history of PSC and/or IBD, have higher rates of CRC. Patients with ALD appear to have a higher than average risk,56, 69 and patients with colon polyps on pre-LT screening are at higher risk for post-LT polyps69; this identifies additional patients requiring further study to optimize colorectal screening recommendations. CRC after LT has been extensively reviewed recently by Nishihori et al.79 In brief, aside from the aforementioned higher risk population, CRC occurs at similar frequency and age as the general population and thus suggests that screening for CRC should be similar in the general LT population (although specific data addressing this do not exist for LT patients).79, 80 Patients with PSC and/or ulcerative colitis are at significantly higher risk for CRC after transplantation, with 5% to 11% of patients (and up to 30% of patients with ulcerative colitis for more than 10 years) developing CRC.71, 72, 79, 81 CRC may present earlier in this population and has been mentioned as early as 9 to 13 months after transplantation; this suggests that screening should start annually after transplantation in this population.71 An age greater than 40 years, colitis for more than 10 years, pancolitis, and colonic dysplasia are risk factors for CRC after transplantation.81 In fact, in patients with 2 or more risk factors, an argument has been made for prophylactic colectomy to lower the risk of CRC and improve long-term survival.79, 81 Annual colonoscopy screening in patients with preexisting PSC and/or IBD is recommended.79, 82 HPV is an established risk factor for anal canal neoplasms and can be identified by screening cytology.74 The use of these methods to identify LT patients requiring routine proctoscopic examinations requires further investigation. Patients with CRC or anal carcinoma can present with asymptomatic anemia, gastrointestinal bleeding, abdominal pain, changes in bowel habits, and/or bowel obstructions. Patients can also be asymptomatic, so screening recommendations have been proposed that are based solely on the aforementioned frequency data because no specific studies have been performed in LT patients. Once patients are diagnosed with CRC, the treatments are similar to those used in the nontransplant population.
Lung cancer occurs at higher rates in LT recipients versus the general population, and this underscores the need for aggressive interventions for smoking cessation.10, 59, 73 The time of presentation frequently ranges from approximately 2 to 6 years after transplantation.59, 83, 84 In a cohort study using the Scientific Registry of Transplant Recipients from 1987 to 2008, LT recipients had a SIR for posttransplant lung cancer of 1.95 (95% CI = 1.86-2.08).10 Lung cancer occurred most commonly in patients with a history of ALD and in smokers (previous or active),56, 59 and this suggests a targeted group of LT recipients to screen more diligently. Most commonly, these are squamous cell carcinomas or adenocarcinomas, but unfortunately, many patients presenting with symptomatic disease have advanced stage malignancies and a worse prognosis.83 Noting a survival advantage for LT recipients with lung cancer in comparison with kidney or heart recipients, 1 study83 suggested that chest X-ray screening may have identified these patients with earlier stage disease. In addition, the greater frequency of adenocarcinoma versus squamous carcinoma may have contributed to improved survival, but the numbers were too small to make any clear associations.
Oropharyngeal/laryngeal malignancies are also more frequent in recipients with ALD and smokers than other LT patients,56, 59 and this finding identifies a subset of patients for which ear, nose, and throat screening should be considered. These malignancies frequently present 1 to 5 years after transplantation.84, 85 No strong evidence exists on screening effects for this malignancy, but patients with malignancies found during routine screening appeared to have better treatment responses and survival than those presenting symptomatically in a small cohort.85 Lower genital tract intraepithelial lesions are increased as early as 6 months after transplantation in female transplant recipients (either liver or kidney) despite negative testing for HPV before transplantation.86 HPV-positive women undergoing renal transplantation have a 14-fold increased risk of cervical cancer, an up to 50-fold increased risk of vulvar cancer, and an up to 100-fold increased risk of anal cancer.66 Only patients with advanced disease will present with symptoms, so screening programs for the general population have been implemented. Very limited data that are specific to LT recipients exist for these cancers (and no data on screening efficacy exist). Until more data suggesting a lower risk in LT recipients are available, annual screening for these malignancies with a Papanicolaou smear and a standard anogenital examination is recommended.
The probability of survival for an LT recipient after a diagnosis of de novo malignancy is dependent on the specific diagnosis, but it is generally worse than the probability for a nontransplant patient with the same cancer. A prospective study using the NIDDK database estimated patient survival to be 62% after 1 year and 47% after 5 years once patients were diagnosed with de novo nonskin solid organ malignancies (the general population estimates for overall survival were 79% at 1 year and 64% at 5 years).56
PREVENTION AND TREATMENT
A number of risk factors for de novo malignancies have been identified, yet many of these risk factors are unmodifiable (eg, recipient age, underlying liver disease, and many viral exposures with oncogenic potential). Thus, regular patient assessments and patient education on the importance of preventative screenings are vital. Because LT recipients should be screened for preexisting malignancies before transplantation as part of a comprehensive ongoing assessment, standard preventative screenings, as summarized in Table 5, can generally commence 1 year after transplantation. All LT recipients should receive counseling on smoking cessation, limitations on alcohol use (complete abstinence in patients with ALD), sun protection and avoidance, regular skin assessments, adherence to routine cancer screening tests, and regular follow-up with their health care providers. In a study assessing the impact of smoking cessation on the risk of cancer development after LT, Herrero et al.59 showed that patients with a smoking history > 20 pack-years who engaged in active smoking after LT or who discontinued smoking < 10 years before LT had a hazard ratio of approximately 20 for the development of smoking-related malignancies.59 In contrast, inactive former smokers had a statistically significant reduction in the risk of smoking-related malignancies, and nonsmokers had an even greater reduction.59 There is also supportive evidence for the idea that the avoidance of excessive sun exposure and the generous use of a sunscreen after LT can help to prevent skin cancer after transplantation.58, 87
Table 5. Screening Suggestions for LT Recipients
Applicable Transplant Recipients
Strength of Evidence
NOTE: This table lists suggestions to be considered by transplant physicians that are based on limited data specific to LT patients (as reviewed in the text). The strength of evidence scoring has been adapted from the US Preventive Services Task Force: (A) good scientific evidence suggests that the benefits of the clinical service substantially outweigh the potential risks; (B) at least fair scientific evidence suggests that the benefits of the clinical service outweigh the potential risks; (C) at least fair scientific evidence suggests benefits provided by the clinical service, but the balance between benefits and risks is too close to make general recommendations; (D) at least fair scientific evidence suggests that the risks of the clinical service outweigh the potential benefits; and (I) scientific evidence is lacking, of poor quality, or conflicting, so the risk-versus-benefit balance cannot be assessed.
Annual dermatological examination
Reduced skin exposure (hats and long sleeves/pants)
EBV status before LT
No screening currently applicable
Annual colonoscopy with random surveillance biopsy
Adherence to the guidelines for an average-risk population
A (general population)
I (LT population)
Annual chest X-ray
Oropharyngeal/ laryngeal cancer
Annual otolaryngology evaluation
Adherence to pelvic examination and Papanicolaou smear guidelines for the general population
A (general population)
B (LT population)
Annual mammography starting at the age of 40 years according to the guidelines for the general population
A (general population)
I (LT population)
Annual prostate-specific antigen testing starting at the age of 50 years according to the guidelines for the general population
C (general population)
I (LT population)
Routine screening tests offer the potential for an earlier diagnosis, which may translate into a higher likelihood of an effective treatment. There is no high-level evidence to support the idea that such protocols make any difference in improving outcomes for de novo malignancies after LT.88 However, in a retrospective analysis of 779 consecutive recipients, Finkenstedt et al.89 suggested improvements in both cancer detection rates and nonskin cancer patient survival with a surveillance protocol [the detection rate increased from 4.9% to 13%, and for nonskin cancers, the median tumor-related survival time improved from 1.2 to 3.3 years (P < 0.05)]. The practicality, cost, and risk of annual computed tomography scans (chest and abdomen) and the increased frequency of posttransplant colonoscopy in this study were not addressed. More data are needed to define the optimal surveillance protocol and to determine to what degree (if any) such a protocol can be individualized to a given patient's particular risk factors. Table 5 outlines screening suggestions to consider that are based on the available clinical data reviewed.
After the diagnosis of a solid organ malignancy, the first-line approach is immunosuppression minimization, particularly during the initial stages while the patient undergoes cancer-specific treatments. Benefits in reduced skin malignancies90 and Kaposi's sarcoma91 have been demonstrated with this approach. Less is known about the impact of dose reduction on other solid organ tumors, but the actual dose reduction is less of a factor for immune function than radiation or chemotherapy treatment is; this suggests a limited role for dose reduction in cancer treatment.92 The balance between the risk of rejection and the benefits of a cancer treatment response needs further study. There are no prospective data to guide clinical decision making on the optimal degree of immunosuppression reduction, but dose reduction to the minimum tolerated dose is suggested. For recipients with preserved graft function, the safety of chemotherapy for the liver is presumed to be similar to that for nontransplant patients, although careful monitoring for drug-drug interactions is important.
Modification of the immunosuppression regimen is another strategy that might positively affect the incidence and outcomes of some de novo cancers after LT. Evidence suggests that calcineurin inhibitors and azathioprine have carcinogenic potential in animal models, including the activation of proto-oncogenes and the distortion of angiogenesis and apoptosis.93, 94 In addition, the increases in transforming growth factor β and interleukin-6 caused by calcineurin inhibitors may contribute to tumor growth and metastatic potential.95 Azathioprine can inhibit DNA repair, and its metabolite 6-thioguanine has been shown in vitro to accumulate in the skin cells; this results in increased oxidant injury with exposure to UV light and thus contributes to the high incidence of skin cancer associated with this agent.96 There are no data showing a definitive risk of malignancy development from exposure to anti–interleukin-2 receptor antibodies and/or mycophenolate mofetil after transplantation.97-99 Antilymphocyte agents, however, have been associated with an increased risk of PTLD.100, 101
The effect of immunosuppression on carcinogenesis appears to be dose-related because the highest incidence of PTLD occurs in the early years after transplantation when the immunosuppression levels are highest, and it appears to respond to the lowering of doses. In addition, the incidence of malignancies appears higher in patients with higher target cyclosporine levels.102, 103 Furthermore, it has also been suggested that quadruple immunosuppression is associated with a higher incidence of malignancies than even triple immunosuppression in the renal transplant population.104
Sirolimus and everolimus, inhibitors of mammalian target of rapamycin (mTOR), are immunosuppressive agents with potential antiproliferative properties. Animal models have shown that mTOR inhibitors are able to suppress tumor growth via multiple mechanisms inhibiting cell proliferation, survival, and angiogenesis.95 This action occurs predominantly through the inhibition of the phosphatidylinositol-3-kinase/AKt signaling pathway and vascular endothelial growth factors.105 In addition, mTOR inhibition may provide synergistic effects to increase cellular apoptosis when it is used in conjunction with other chemotherapeutic agents to treat malignancy.106, 107 The modest antitumor effects of the current mTOR inhibitors have been reviewed recently.108 Sirolimus has a unique side-effect profile and is not widely used after LT; this limits observational data specific to the LT population.
There is controversy surrounding the effectiveness of mTOR inhibitors in the prevention of malignancies because the antitumor effects may be tumor-specific and dose-dependent. Clinically, there is some evidence for a reduced frequency of malignancies (with most data on skin malignancies) after renal transplantation in patients receiving sirolimus versus standard calcineurin inhibitor immunosuppression regimens.109 A recent open-labeled randomized controlled trial of 830 renal transplant patients with 2 years of follow-up showed a reduction in skin cancer rates but no significant reduction in other malignancies.110 Similarly, in a safety and efficacy trial comparing sirolimus-based regimens to tacrolimus and mycophenolate mofetil regimens in adult kidney recipients, there was no difference in the malignancy rates of the 2 groups.111 Trials of sirolimus in LT recipients have reported a high discontinuation rate (∼46%) because of severe adverse events, including rejection, that perhaps counter the risk/benefit balance.112 Specific effects of mTOR inhibitors on malignancy rates after LT are not known, but they presumably would mirror the effects in the renal transplant literature. Even the literature on mTOR inhibition and the recurrence of hepatocellular carcinoma after LT is mixed, with a low level of evidence supporting or dissuading its use.113 There may be a role for mTOR inhibitors in the treatment of de novo malignancies after transplantation, but data on their effectiveness on the long-term outcomes of these patients are not yet known. One specific malignancy that may benefit from mTOR inhibitors is Kaposi's sarcoma, but data are limited to case reports.114
Without further randomized controlled data, current recommendations support the minimization of immunosuppression to the lowest tolerable level. The replacement of azathioprine by mycophenolate mofetil potentially reduces the risk of skin cancer. The risk-to-benefit ratio of calcineurin inhibitor-free immunosuppression in favor of mTOR inhibitor–based immunosuppression is unclear. Further studies of the benefits of mTOR inhibitors in LT recipients, with a potential focus on patients with a higher risk of de novo malignancies after transplantation (eg, older patients, patients with alcohol-related disease, and patients with a history of smoking), are needed. Many of these high-risk factors are also associated with a lower rate of rejection and thus may make this kind of study more attractive and feasible.
De novo malignancies are leading causes of morbidity and mortality in LT recipients. Targeted efforts to reduce the burden of posttransplant malignancies may improve long-term outcomes. Lymphoproliferative and solid organ malignancies have numerous risk factors, including recipient characteristics, the underlying liver disease, immunosuppression, and environmental exposures. Future studies assessing reduced and/or alternative immunosuppression regimens and protocolized screening and multidisciplinary care pathways are required. Multicenter consortiums should be established to facilitate the growth of high-quality research in this area.