SEARCH

SEARCH BY CITATION

Keywords:

  • Cancer;
  • head and neck;
  • squamous cell carcinoma;
  • transplant;
  • immunosuppression.

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY

Introduction There is an increased incidence of cancer in patients after organ transplantation. We reviewed a large series of cardiothoracic transplant recipients to determine the incidence and natural history of head and neck malignancy.

Methods A total of 1069 heart (n = 855), heart/lung (n = 111), and lung (n = 103) transplants were performed at Stanford University from January 1968 to February 1998. Demographic data, risk factors, and disease course were evaluated in patients who developed cancer. The mean length of follow-up was 8.9 ± 5.2 years.

Results One hundred twenty patients (11.2%) developed 547 non-lymphomatous malignancies. The mean number of malignancies per cancer patient was 4.6. The average time from transplantation to development of cancer was 63.1 months. A total of 50.5% of malignancies presented in the head and neck; 96.4% of these were cutaneous in origin and 3.6% were noncutaneous. Of cutaneous malignancies, 79.3% were squamous cell carcinoma and 15.9% were basal cell carcinoma. Cutaneous malignancies most commonly presented on the scalp, cheek, lip, and neck. Noncutaneous malignancies involved the oral cavity (5), thyroid (4), and parotid (1). Thirteen percent of cutaneous head and neck cancers behaved aggressively, requiring e-tensive management including radical surgery, radiation, and/or chemotherapy. A total of 34.2% of cancer patients developed metastases and 54.9% of cancer patients died as a direct result of cancer. A total of 68% of cancer patients were smokers and 23.8% had significant alcohol use.

Conclusion Transplant recipients have an increased incidence of cancer presenting in the head and neck. Malignancies in transplant patients behave more aggressively than in the general population. Recognition of this aggressive biological behavior and heightened cancer surveillance should result in improved outcomes.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY

The increased incidence of cancer in transplant recipients has been well documented. 1–8 Cutaneous squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) are the most prevalent of these malignancies, followed by non-Hodgkin's lymphoma (NHL). One of the most common locations for solid malignancies in transplant recipients is the head and neck, 2,4,5,8–11 yet there are no studies focusing on the incidence and behavior of head and neck malignancies in the cardiothoracic transplant population.

We therefore sought to characterize the incidence, behavior, and risk factors for development of all de novo solid head and neck tumors after heart, heart/lung, and lung transplantation. A pioneering institution in the field of transplantation, Stanford University ranks as one of the busiest transplant centers in the world, with over 1000 cardiothoracic transplants performed over the past 30 years—providing adequate data to answer the questions posed.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY

Between January 1968 and February 1998, a total of 1069 cardiothoracic transplants were performed at Stanford University, including 855 heart, 111 heart/lung, and 103 lung transplants. Patients who developed malignancies after transplantation were identified from the transplant tumor registry. The charts and pathology records of these patients were retrospectively reviewed. Clinical data evaluated included age, sex, race, length of follow-up, pretransplant diagnosis, immunosuppressive regimen, number of rejection episodes, number of re-transplants or addition of a kidney transplant, interval between date of transplantation and diagnosis of a malignancy, type of malignancy, location of malignancy, progression to metastases, cause of death, prior history of cancer, family history of cancer, tobacco use, alcohol use, and infection with Epstein-Barr virus, human papilloma virus, HIV, herpes simplex virus, cytomegalovirus, or hepatitis C virus.

In some patients with multiple cutaneous malignancies, the physicians' records, operative notes, and pathology reports were not adequate for determining the exact number of distinct lesions. For example, an operative note might read “excision of multiple SCCs from forehead.” When the precise number of lesions was ambiguous, the patient was noted to have “at least one” of the specific malignancies and one lesion was added to the cumulative total. Therefore, this approach likely underestimates to some extent the total number of malignancies in the study group.

Specific criteria had to be met in order for tumors to be considered posttransplant de novo solid malignancies. Tumors that were diagnosed clinically, radiographically, or pathologically before transplantation were excluded from the study. Benign tumors, in situ carcinomas, premalignant lesions, and patients with posttransplant lymphoproliferative disease (PTLD) only were also excluded. Alcohol consumption was divided into three subgroups: none; one or two drinks per day (moderate consumption); and three or more drinks per day (heavy consumption).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY

Patient Characteristics

One hundred eighty-one patients (16.9%) developed malignancies after cardiothoracic transplantation. One hundred twenty of these developed at least one solid malignancy; the remaining 61 patients developed lymphoproliferative malignancies only. Of the 120 patients with solid malignancies, 106 were men and 14 were women. One hundred fifteen were white, and there were two blacks, two Hispanics, and one Asian. The mean age at the time of transplantation was 45.7 ± 13.3 years. The mean length of follow-up was 8.9 ± 5.2 years.

Pretransplant diagnosis was determined for 100 patients in the solid malignancy group. They included 40 patients with coronary artery disease, 36 with idiopathic cardiomyopathy, 4 with Eisenmenger's syndrome, 4 with familial cardiomyopathy, 3 with valvular disease, 3 with primary pulmonary hypertension, 2 with viral cardiomyopathy, 2 with postpartum cardiomyopathy, 2 with pulmonary fibrosis, 1 with congenital heart disease, and 1 with bronchiectasis.

Immunosuppression and Posttransplant Course

Patients were treated with immunosuppressive agents based on the transplant protocols in use at the time. These agents (and the number of patients treated) included prednisone (116), cyclosporin (96), azathioprine (112), OKT3 (50), antithymocyte globulin (39), mycophenolate mofetil (9), Neoral (8), total lymphocyte irradiation (6), methotrexate (4), cyclophosphamide (4), and FK-506 (3).

There were a total of 249 rejection episodes, representing a mean of 2.1 ± 2.1 episodes per patient (range, 0–10). Transplants were performed again in 6 patients (5.0%) and 5 patients (4.1%) received kidney transplants in addition to their heart transplant.

Solid Malignancies

A total of 547 solid malignancies occurred in the cohort of 120 cancer patients, representing a mean (± SD) of 4.6 ± 5.5 malignancies per cancer patient (range, 1–32) (Fig. 1). The interval from transplant to diagnosis of cancer ranged from 1 month to 22.4 years, with a mean of 63.1 ± 56.1 months. The first malignancy type to develop was cutaneous SCC (46.1%), BCC (20%), genitourinary cancer (11.3%), gastrointestinal cancer (7.0%), lung cancer (4.3%), noncutaneous head and neck cancer (2.6%), melanoma (2.6%), PTLD (2.6%), breast cancer (1.7%), Kaposi's sarcoma (0.9%), central nervous system cancer (0.9%), and undetermined (4.2%) (Fig. 2). The first malignancy presented in the head and neck region in 64 (56.1%) of 114 of the patients in whom this was determined.

thumbnail image

Figure Fig. 1.. Pie graph showing the prevalence of various types of malignancy in the posttransplant population. Cutaneous malignancies are by far the most common. White = cutaneous head and neck; black = noncutaneous head and neck; dark grey = cutaneous non–head and neck; light grey = other non–head and neck.

Download figure to PowerPoint

thumbnail image

Figure Fig. 2.. Number of patients with a given malignancy as their first posttransplant malignancy. GI = gastrointestinal; GU = genitourinary; BCC = basal cell carcinoma; SCC = squamous cell carcinoma.

Download figure to PowerPoint

Head and Neck Malignancies (Cutaneous and Noncutaneous)

Of the 547 solid malignancies, 276 (50.5%) occurred in the head and neck in 80 patients (66.7%). The total number of malignancies presenting in the head and neck per cancer patient ranged from 0 to 21, with a mean (± SD) of 2.3 ± 3.3. Two hundred and sixty-six (96.4%) of these were cutaneous in origin and 10 (3.6%) were noncutaneous.

The 270 cutaneous tumors of the head and neck were SCC (n = 214, 79.3%), BCC (n = 43, 15.9%), Kaposi's sarcoma (n = 7, 2.6%), melanoma (n = 1, 0.4%), porocarcinoma (n = 1, 0.4%), and 4 (1.5%) with unknown histology. Ten locations were involved by these malignancies (Fig. 3). Cutaneous malignancies invaded the parotid gland in four (3.3%) patients. The ratio of BCC to SCC for the head and neck was 1 to 4.98.

thumbnail image

Figure Fig. 3.. Distribution of posttransplant cutaneous head and neck malignancies as a percentage of all posttransplant cutaneous head and neck malignancies.

Download figure to PowerPoint

The 10 noncutaneous head and neck malignancies occurred in 6 patients (5.0%) and involved the oral cavity (n = 5, 1.9%), the thyroid gland (n = 4, 1.5%), and the parotid gland (n = 1, 0.4%). Malignancies of the head and neck (cutaneous or noncutaneous) were classified as “aggressive” when they were associated with metastatic disease, the need for extensive therapeutic management (radiation therapy, chemotherapy, or radical surgical intervention), or when they were the direct cause of death. Twenty-six (32.5%) of the patients with solid malignancies of the head and neck (2.4% of all cardiothoracic transplant recipients) were deemed to have aggressive malignancies. There were a total of 30 aggressive cutaneous tumors and 10 aggressive noncutaneous tumors. Overall, 13.0% of the cutaneous head and neck malignancies were aggressive.

Other Malignancies

Two hundred fifty-seven solid malignancies occurred in 87 patients in areas other than the head and neck. Of these, 214 (83.3%) were cutaneous and 43 (16.7%) were noncutaneous. Non–head and neck cutaneous malignancies occurred in 57 patients (5.3% of all transplant recipients), representing a mean (± SD) of 3.8 ± 5.3 lesions per patient (range, 1–31). The overall ratio of BCC to SCC for all locations was 1 to 2.53. The 43 noncutaneous solid malignancies occurred in 38 patients (Fig. 4).

thumbnail image

Figure Fig. 4.. Site and number of non–head and neck posttransplant malignancies. GI = gastrointestinal; GU = genitourinary.

Download figure to PowerPoint

For 14 patients, there was reference to one or more cutaneous malignancy without identification of location. These were scored as “at least one cutaneous malignancy” and account for the remaining 14 malignancies (5.4%) of the 547 total malignancies reported in this study. PTLD developed in 10 patients (8.3%) in addition to their solid malignancies. Seven of these patients had at least one solid malignancy of the head and neck.

Disease Course

Metastatic disease developed in 41 (34.2%) of the 120 patients with solid malignancies. The malignancies that progressed to metastasis (indicated as number of patients) were breast cancer (2 of 2, 100%), noncutaneous head and neck cancer (5 of 6, 83.3%), melanoma (4 of 5, 80%), lung cancer (5 of 9, 55.6%), gastrointestinal cancer (6 of 12, 50%), genitourinary cancer (6 of 19, 31.6%), Kaposi's sarcoma (1 of 2, 50%), cutaneous SCC (10 of 78, 12.8%), and BCC (1 of 52, 1.9%).

A total of 71 cancer patients (59.2%) were deceased at the time of the study; 39 (54.9%) died as a direct result of cancer. The posttransplant life span of these 71 patients ranged from 3.3 months to 20.5 years, with a mean of 8.0 ± 4.9 years. In eight patients, the type of cancer that resulted in death was not reported in the records; the type of cancer resulting in death in the remaining 31 patients is listed in Figure 5.

thumbnail image

Figure Fig. 5.. Bar graph depicting the percentage of posttransplant cancer-related deaths due to each type of malignancy. GI = gastrointestinal; GU = genitourinary; BCC = basal cell carcinoma; SCC = cutaneous squamous cell carcinoma; PTLD = posttransplant lymphoproliferative disorder.

Download figure to PowerPoint

Risk Factors

Ten patients had a prior history of malignancy (9.3% of the 108 cancer patients for whom data were available). A family history of cancer was present in 35 of 93 patients (37.6%) for whom data were available. Data regarding tobacco use was available for 103 patients; 70 (68.0%) had a positive history. For 9 of these patients, the amount of usage was not indicated; the remaining 61 patients reported a mean (± SD) of 46 ± 28 pack years of usage (packs per day multiplied by years of smoking) (range, 5–160). Of the 99 patients for whom an alcohol consumption history was documented, 61 (76.3%) were moderate drinkers (1–2 drinks per day), and 19 (23.8%) were heavy drinkers (≥ 3 drinks per day).

Posttransplant viral infections were documented in 142 patients and included herpes simplex virus (n = 68, 56.7%), cytomegalovirus (n = 61, 50.8%), Epstein-Barr virus (n = 7, 5.8%), and hepatitis C virus (n = 6, 5.0%). There were no documented cases of HIV or human papilloma virus infections.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY

More than 165,000 solid organ transplants were performed in the United States between 1988 and 1997. 12 These transplant recipients suffer significant treatment-related co-morbidities that include long-term toxic effects of medications, susceptibility to opportunistic fungal and viral infections, graft rejection, and neoplasms. In 1972 Penn et al. 13 reported that transplant recipients had approximately a 6% risk of developing de novo malignant tumors. This increased risk was 100 times greater than that of the matched general population. The studies that followed focused on either cutaneous malignancies or PTLD. In 1981 Disler et al. 14 showed that transplant recipients with cutaneous SCC had a greater risk of developing metastatic disease compared with the general population. Five years later, a study reported by Gupta et al. 9 revealed that 74% of cutaneous SCC and BCC in the renal transplant population presented on the face, head, and neck. They also determined that the rate of SCC occurrence was 18.4 times that of the general population. That same year, data from the Cincinnati Transplant Tumor Registry 15 showed that lymph node metastases occurred in approximately 8% of patients with posttransplant cutaneous malignancies and that 7% of tumor-related deaths were due to posttransplant cutaneous malignancies. Studies since that time have confirmed that a transplant patient's risk of developing cutaneous malignancies increases over time and that the most common region in which cutaneous malignancies develop is the head and neck. 2,5,8,10,11,16 A recent study reports that posttransplant patients with aggressive cutaneous malignancies have significantly more cephalic lesions that require more cosmetic and reconstructive surgery and radiation therapy than patients with nonaggressive lesions. 16

Very few data have been generated regarding the occurrence of noncutaneous solid malignancies after transplantation. In 1995 two studies suggested that the incidence of posttransplant noncutaneous solid malignancy ranges from 3% to 6%. 6,7 Head and neck cancers accounted for 10% to 16% of all noncutaneous solid malignancies. These studies were conducted with relatively modest sample sizes of 608 patients 7 to 633 patients. 6

The study presented here is the largest, single-center posttransplant evaluation for occurrence of solid malignancies. We assessed the incidence and behavior of cutaneous and noncutaneous solid malignancies, which occurred in 11.2% of the patients in this group. More than half of the first malignancies after transplantation presented in the head and neck region. The majority of head and neck malignancies were cutaneous in origin, whereas only 3.6% of head and neck malignancies were noncutaneous. Interestingly, 13.0% of all head and neck malignancies were aggressive; 32.5% of patients with a cutaneous malignancy had one that was aggressive. It is therefore important for head and neck surgeons who participate in the care of patients with a history of organ transplant to be vigilant in their efforts to prevent, screen for, and aggressively treat premalignant conditions. An additional finding in our analysis was the reversal of the usual ratio of BCC to SCC in cutaneous cancers. Normally this ratio is between 2:1 and 5:1; in our study it was 1:2.5 overall and 1:5 in the head and neck.

Another compelling statistic we encountered is that metastatic disease developed in one third of posttransplant cancer patients. Noncutaneous head and neck malignancies metastasized in five of six patients. PTLD was the most common cause of cancer-related death but was followed by cutaneous SCC. Noncutaneous head and neck malignancies were the cause of death of 9.7% of patients with a known cancer-related death.

The most widely cited explanation for the increased risk of cancer after transplantation is that immunosuppressive medications impair the tumor surveillance mechanism of lymphocytes, thereby disrupting the balance between tumorigenesis and tumor lysis in the body. Azathioprine, a mutagenic purine analogue, is now thought to act as a synergistic agent with ultraviolet radiation, potentially increasing the risk of skin cancers. 17–19 Ultraviolet radiation has been shown to alter the p53 suppressor gene, located on chromosome 17 (17p13), in cutaneous cells, thereby predisposing them to neoplasia. 20 Ultraviolet radiation also affects the cells of Langerhans, 21 and may effect T-suppressor and dendritic cells, 22 impairing the cutaneous surveillance system. 23 As a result, ultraviolet light may act as a local and systemic immunosuppressant, leading to skin tumors even in nonexposed areas. 24 Transplant recipients should therefore do everything possible to avoid sun exposure and physicians caring for these patients should provide screening opportunities that allow for the early detection and subsequent management of worrisome lesions.

Viral oncogenesis has been strongly supported by the association of Epstein-Barr virus in non-Hodgkin's lymphoma 25 and nasopharyngeal carcinoma. 26 Human papilloma virus is associated with alterations of the p53 tumor suppressor gene. 27 Although human papilloma virus has been implicated as a potential cause of skin cancer, it is unclear if the virus plays a substantial role in cutaneous neoplasms in transplant patients. 4 Interestingly, we found that none of the transplant recipients with solid malignancies had a documented history of human papilloma virus, lending further doubt to the contribution of viral oncogenesis in posttransplant malignancy.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY

De novo solid malignancies occur at a high rate in postcardiothoracic transplant patients. The majority of the solid malignancies occur in the head and neck. The solid malignancies, including cutaneous tumors, in transplant recipients behave in a much more aggressive fashion than those in the nontransplant population. Head and neck surgeons should maintain a high level of cancer surveillance in postcardiothoracic transplant patients. When malignancies occur, they should be aggressively treated.

BIBLIOGRAPHY

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. BIBLIOGRAPHY