Financial disclosures: Fees from lectures for Roche, Novartis and Wyeth.
Funding sources: CIBERehd is funded by the Instituto de Salud Carlos III.
J Ignacio Herrero, Liver Unit, Clínica Universidad de Navarra, Av Pio XII, 36, 31008 Pamplona, Spain. Email: firstname.lastname@example.org
De novo malignancies are frequent complications after liver transplantation. They are one of the leading causes of late death. Some authors have reported promising results following implementation of extensive cancer surveillance programs. Screening protocols might benefit patients by providing a diagnosis at an earlier stage when tumors may be cured. These protocols should be based on the specific risk factors of every patient. Unfortunately, the scientific evidence supporting screening protocols is still very weak both in the general population and in the transplant patients. On this basis, there is not enough evidence to recommend routine screening for all liver transplant recipients, apart from the recommendations accepted for the general population. Multicenter studies in selected groups of patients at high risk for malignancy may be the only way of defining the potential benefit of screening programs post-transplantation.
One of the most important complications of immunosuppressive therapy for liver transplantation is the development of neoplasia.1 Immunosuppressive therapy inhibits immune surveillance against malignant cells and a variety of viruses with oncogenic potential. Moreover, certain immunosuppressive drugs may have a direct oncogenic effect. Consequently, transplant recipients have a high incidence of neoplasia: more than one-third of transplant recipients develop a tumor at 10 years post-transplantation.1 Several authors have shown that the risk of malignancy is higher in transplant recipients than in the sex- and age-matched general population.2–6 The risk of death due to malignancy is significantly higher in liver transplant recipients despite the fact that they are more closely followed than the general population.2,3,7
Some indications for liver transplantation are associated with cancer. Colorectal cancer is common in patients with primary sclerosing cholangitis and ulcerative colitis,8 while patients with alcoholic liver disease have a higher risk of neoplasia, both because the consumption of alcoholic beverages increases the risk of several malignancies,9 and because high alcohol intake is frequently associated with smoking, a key risk factor associated with some of the most prevalent malignancies, such as lung, head and neck, esophagus and urinary tract carcinomas.10 Last but not least, patients transplanted in the last decade are older than their predecessors, and therefore are at greater risk of neoplasia. As a result, malignancy is a very important cause of mortality after liver transplantation11 (Fig. 1). In fact, it is the leading cause of late death in some series,1,12 accounting for more than 25% of mortality overall, and up to 40% of late deaths.1
The risk of malignancy is 2–4 times higher in liver transplant recipients than in the age- and sex-matched general population.1 This increase is not uniform (Table 1). It is especially high (more than 10-fold) for skin cancer,6,17,18 non-Hodgkin's lymphoma,6 Kaposi's sarcoma17 and uterine cervical cancer. Most of these tumors are related to viral infections. The risk of some common malignancies, such as breast and prostate adenocarcinomas is not increased, while the risk of others, such as head and neck cancer, lung, urological or colorectal carcinomas is only moderately increased. 1
Table 1. Relative risk of neoplasia in liver transplant recipients, in comparison with a sex-matched, and age-matched population (references 2–6,13–16 2–6, 20, 29, 51, 54)
Immunosuppression. The high risk of malignancy after liver transplantation may be attributed to immunosuppression, but there is not a close correlation between the development of neoplasia and specific protocol of immunosuppression. There are few exceptions: some evidences suggest that the intensity of immunosuppression is related to the risk of neoplasia.20,21 Also, there is a close association between therapy with antilymphocyte globulins and the risk of post-transplant lymphoproliferative disease.22 Retrospective analyses have revealed a lower incidence of malignancy in kidney transplant recipients treated with mammalian target of rapamycin (mTOR) inhibitors, such as sirolimus or everolimus. This protective effect has been attributed to their anti-proliferative properties.23
Age. Children and young adults have a high risk of lymphoma because they are frequently seronegative for Epstein–Barr virus (EBV), and transplantation of an organ from an EBV seropositive donor to an EBV seronegative recipient is associated with a high risk of lymphoma.13,24 Apart from this increased risk of lymphoma, the risk of malignancy increases with age,2,3,25,26 like in the general population. The 10-year risk of malignancy in patients older than 60 nearly doubles the risk of patients below this age (52% vs 28%).1 In fact, the high risk of malignancy is one of the most important causes of impaired survival after transplantation in older recipients.27 This notwithstanding, the relative risk of neoplasia, as compared with age- and sex-matched controls, is much higher in younger transplant recipients;5 patients younger than 40 years have a risk 4 times greater than the general population of the same age, while the risk is only two times greater for people older than 40 years.5
Other risk factors. Smoking is an independent risk factor for non-cutaneous malignancy in liver transplant recipients.2,26 As in the general population, smoking is related to a higher risk of esophageal,28 head and neck, lung29 and urological (other than prostate) carcinomas.10 In fact, the risk of these tumors is close to zero in non-smoking patients (Fig. 2). Alcohol is also a potent carcinogen (mainly in smokers),9 and it has also been related to higher incidences of oropharyngeal, esophageal and pulmonary carcinomas.19 Thus, patients transplanted for alcoholic liver disease have a higher risk of non-skin neoplasia.2,26
Apart from the previously mentioned association between EBV infection and post-transplant lymphoproliferative disease, other risk factors include use of antilymphocyte globulins, as well as cytomegalovirus and hepatitis C viral infections.24
Squamous-cell and basal-cell cutaneous carcinomas are mainly related to fair skin type and cumulative sun radiation exposure:18 patients with skin type I or II (i.e. skin that is usually burned after sun exposure) who have been exposed to considerable solar radiation have a 5-year risk of non-melanoma skin cancer higher than 50%.18 Kaposi's sarcoma is related to infection by human herpesvirus type 8, but the absolute risk of this neoplasia is low.17
Patients with ulcerative colitis (frequently associated with primary sclerosing cholangitis) have a high risk of colorectal cancer,5 especially patients with colitis of long duration. The potential correlation between viral infections (EBV and JC virus) and colorectal neoplasia of renal and liver transplant recipients has been suggested.30,31 This association could better explain that transplant recipients without ulcerative colitis also have a high incidence of colorectal carcinoma.14,30
Screening for specific malignancies
The aim of cancer screening programs is the detection of malignancy at an early stage, when potentially curative treatment is possible. Thus, cancer screening in liver transplant recipients seems justified, as it is one of their leading causes of death. To evaluate if a screening program is cost-effective, we must balance the benefit of early diagnosis against the potential harms of screening. The following factors must be evaluated: prevalence of neoplasia, expected survival of the population, sensitivity and specificity of a given screening test, possibility of curative therapy, mortality, potential harm (directly or through false positive or negative results), and cost. Unfortunately, cancer screening in liver transplant recipients is not supported by current scientific data. Recommendations for screening are extrapolated from general guidelines, but these recommendations are not directly applicable to liver transplant recipients for several conflicting reasons. Liver transplant recipients have a higher incidence of malignancy, so an early diagnosis could potentially benefit more patients. Unfortunately, the potential benefit of early diagnosis in liver transplant recipients might be overestimated, because the life expectancy of these patients is usually lower than the general population. Furthermore, liver transplant patients frequently suffer from comorbidities, which may curtail treatment with curative intent.
Skin cancer. Skin cancers are the most common type of de novo neoplasia after liver transplantation. Transplant recipients have a higher incidence of skin cancer than the general population: close to 20% of liver transplant recipients have been diagnosed with skin cancer 5 years after transplantation.18 Fortunately, most of these cancers are basal cell and squamous cell carcinomas and they do not limit survival.18 Melanoma is associated with a worse prognosis than the aforementioned carcinomas, but fortunately, its incidence is very low.17
The United States Preventive Services Task Force's (USPSTF) most recent recommendations concluded that current evidence does not support screening for skin cancer in the general population.32 The 2009 Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guidelines suggest that physical examination by the patient or a health professional might be beneficial for kidney transplant recipients,33 but it was considered that the quality of evidence was very low. More strict recommendations, given by the American Society of Transplantation for renal transplant patients, included monthly self-examination of the skin and examination of the skin by a physician at least every year.34 The recommendation for liver transplant recipients could be the same as for kidney transplant patients, as it seems reasonable that their risks should be similar. This recommendation may be reinforced in patients with pale skin and a high exposure to sun radiation,18 or to those patients who had been previously diagnosed with another cutaneous malignancy.
Post-transplant lymphoproliferative disease. Monitoring EBV DNA levels may be of value in the early diagnosis of lymphoma.35 Unfortunately, an exact cut-off value of EBV DNA critical for the development of lymphoma has not been identified. Patients with a high risk of lymphoma, according to their EBV DNA levels could benefit from antiviral therapy36 and/or a reduction in immunosuppression,37 but the results of uncontrolled studies, while promising, are insufficient to support any recommendation. There is also a lack of evidence about the duration of screening. Most lymphomas are diagnosed in the first year after transplantation,35 but most centers monitor EBV DNA for longer periods of time.36,38
A recent retrospective study found a beneficial effect of EBV monitoring. Pediatric transplant patients routinely screened for EBV DNA levels showed a reduction in advanced lymphoproliferative disease when compared to historical controls. No deaths due to lymphoma were reported in the screened population.38
Lung cancer. Lung cancer screening with chest radiography and sputum cytology have failed to demonstrate an improvement in mortality. More recently, low-dose computed tomography scanning (LDCT) has shown promising results. LDCT is able to detect a higher proportion of lung cancers than chest radiography.39 In the International Early Lung Cancer Action Project, 85% of lung cancers detected by computed tomography screening were stage I.40 Very recently, the National Lung Screening Trial has found 20% fewer lung cancer deaths in heavy smokers who were screened with LDCT than in patients who were screened with chest radiography.41
Despite these encouraging results, LDCT screening remains controversial because of its high cost and false positive results, and current guidelines do not recommend the use of computed tomography to screen for lung cancer.42
Head and neck and esophagus cancer. Despite the fact that there are no current recommendations for or against screening of head and neck cancer for the general population, this is one of the most frequent malignancies leading to death after liver transplantation.25 Esophageal neoplasia is less frequent, but its prognosis is very poor; its main important risk factor is alcohol abuse.28 Consequently, several liver transplant centers have incorporated head and neck and esophageal cancer screening protocols for smokers, especially if they have a history of alcohol abuse.
Colorectal cancer. Recent guidelines recommend screening for colorectal cancer in the general population between 50 and 75 years of age. Screening of older patients should be individualized, and those aged 85 or more should not be screened for colorectal cancer.43 Screening should be based on fecal occult blood testing, sigmoidoscopy or colonoscopy. The American College of Gastroenterology recommends colonoscopy every 10 years as the preferred technique,44 but recent investigations question whether colonoscopy is sensitive enough.45 As transplant recipients have a moderately increased risk of colon cancer, some authors suggest that they should undergo screening for colorectal cancer, according to recommendations given for the general population:46 screening in this population might include fecal occult-blood testing performed every year, sigmoidoscopy every 5 years, or colonoscopy every 10 years.47 A recent study suggested that colorectal cancer screening using occult blood testing could be cost-effective in kidney transplant recipients.48
Patients transplanted for primary sclerosing cholangitis who have ulcerative colitis require special attention. The incidence of colorectal cancer is close to 15% at 5 years in this group of patients,8 but it is not clear whether it is only due to long-term ulcerative colitis or whether immunosuppressive therapy increases this risk. These patients require close surveillance. The American Association for the Study of Liver Diseases recommends that patients transplanted for primary sclerosing cholangitis who have ulcerative colitis should undergo annual surveillance with colonoscopy.49
Urologic malignancies. According to current guidelines, routine screening for prostate cancer, including digital rectal examination or testing for prostate-specific antigen, is not supported by the existing evidence, because screening does not seem to improve mortality.50 As liver transplant recipients do not have an increased risk of prostate cancer and their expected survival is lower than the general population, routine prostate cancer screening of liver transplant recipients should not be recommended.
Some studies have shown encouraging results in the screening of bladder cancer in heavy smokers.51 Nonetheless, the most recent recommendations from the USPSTF do not support routine screening for bladder cancer in the general population, because of its low prevalence and invasiveness.52 Liver transplant recipients could benefit from bladder cancer screening, because they have a higher incidence of urological malignancies, which possibly have a more aggressive behavior in liver transplant recipients than in the general population.15
Breast cancer. General recommendations for breast cancer screening include a mammography every 2 years in women aged 50–74.53 As liver transplant recipients do not have an increased risk of breast cancer, screening recommendations should probably be the same for this patient population.
Recommendations of screening in liver transplantation
Some authors have recommended cancer screening in liver transplant recipients.13,15,16,54 There is no agreement between them and a lack of evidence supporting specific testing protocols. Most of these recommendations focus on the most frequent malignancies diagnosed in liver transplant recipients: cutaneous, colorectal, pulmonary, head and neck, and urological cancers. Table 2 shows a potential surveillance program. This proposal is not based on scientific evidence: it takes recommendations accepted for the general population and the author's personal recommendations.
Table 2. Potential suggestion of malignancy surveillance in liver transplant recipients
Population at risk
More frequent evaluation could be recommended in patients with fair skin type, high exposure to sun radiation or history of skin cancer.
CT, computed tomography; EBV, Epstein–Barr virus; ENT, ear, nose and throat.
Results of malignancy screening programs in liver transplant recipients
Two recent publications suggest that an extensive cancer surveillance program is associated with improved survival. Finkenstedt et al.,16 showed that patients undergoing intensified cancer surveillance had improved survival after the diagnosis of malignancy than historical controls screened by annual chest radiography and abdominal ultrasound alone. The intensified surveillance protocol included annual chest and abdominal computed tomography scans, urological, dermatological and gynecological evaluations and colonoscopy every 5 years.
We have previously published our experience in Pamplona15 with intensified cancer surveillance, showing that despite incomplete adherence to screening protocols, all transplant patients at our center diagnosed with carcinoma as part of a surveillance protocol, underwent curative treatment. Thus, they had significantly better survival than patients with symptomatic or incidental carcinomas.
Since these programs are still not well defined, their encouraging results must be taken with caution.
Malignancy is more frequent in transplant recipients than in the general population, and one of the leading causes of late mortality after liver transplantation. Several authors have suggested different programs of neoplasia surveillance. Intensive screening programs may detect malignancies early enough to allow for curative treatment, but evidence to support these recommendations is weak.
Unfortunately, extrapolations from the general population or from kidney transplant recipients are arbitrary, because their epidemiology and prognosis are different. On this basis, there is not enough evidence to support extensive surveillance protocols for the early diagnosis of cancer in liver transplant recipients. In general, established screening practices for breast, colonic, and cervical cancer that are accepted for non-transplant individuals should be maintained. For other malignancies that are more frequent in transplant recipients, the potential role of screening needs to be investigated, in order to improve survival in the patients who develop malignancies. Future studies must focus on a better definition of specific risk factors for the development of each neoplasia and the potential risk or benefit of every screening test. If we want to obtain evidence-based conclusions about the potential benefit of surveillance tests and their cost-effectiveness, multicenter studies should be conducted.
We wish to thank Luis M Seijo, MD for the critical review and English proof editing of the manuscript.