SEARCH

SEARCH BY CITATION

Keywords:

  • Graft survival;
  • immunosuppression;
  • multivariate survival analysis;
  • patient survival;
  • post-transplant lymphoproliferative disorder;
  • renal transplant

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Significant mortality is associated with post-transplant lymphoproliferative disorder (PTLD) in kidney transplant recipients (KTX). Univariate/multivariate risk factor survival analysis of US PTLD KTX reported to Israel Penn International Transplant Tumor Registry from November 1968 to January 2000 was performed. PTLD presented 18 (median) (range 1–310) months in 402 KTX. Death rates were greater for those diagnosed within 6 months (64%) versus beyond 6 months (54%, p = 0.04). No differences in death risk for gender, race, immunosuppression, EBV, B or T cell positivity were identified. Death risk increased for multiple versus single sites (73% vs. 53%, hazards ratio (HR) 1.4). A 1-year increase in age increased HR for death by 2%. Surgery was associated with increased survival (55% vs. 0% without surgery) (p < 0.0001). Patients with allograft involvement, treated with transplant nephrectomy alone (n = 20), had 80% survival versus 53% without allograft removal (n = 15) (p < 0.001). Overall survival was 69% for allograft involvement alone versus 36% for other organ involvement plus allograft (n = 19 alive) (p < 0.0001). Death risk was greater for multiple site PTLD and increasing age, and risks were additive. Univariate analysis identified increased death risk for those not receiving surgery, particularly allograft involvement alone.


Abbreviations: 
ALG

Anti-lymphocyte globulin preparation;

ATG

Anti-thymocyte globulin preparation;

CNS

Central nervous system;

EBV

Epstein–Barr virus;

IPITTR

Israel Penn International Transplant Tumor Registry;

ISD

Immunosuppression discontinuation;

ISR

Immunosuppression reduction;

ISRD

Immunosuppression reduction or discontinuation;

KTX

Kidney transplant recipients;

OKT3

Muromonab monoclonal antibody preparation;

PTLD

Post-transplant lymphoproliferative disorder

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Post-transplant lymphoproliferative disorder (PTLD) was recognized over three decades ago (1) and has been noted throughout all immunosuppressive eras (2–4). PTLD affects less than 1% of kidney transplant recipients (KTX) but results in death in up to 44% of patients. Recently, an increase in the incidence of PTLD has been reported and has suggested that calcineurin inhibitors may be responsible for the increase (5). Previous literature reviews have indicated that various factors, including number of site involved and Epstein–Barr virus (EBV) status may predict survival in patients with PTLD (6).

The Israel Penn International Transplant Tumor Registry (IPITTR) is a voluntary registry (reporting is not mandatory), which has collected data on malignancies in transplant patients from 1968 to 2004. Multivariate risk factor analyses for PTLD outcomes and survival have been limited to relatively small experiences (7–9) and previous IPITTR PTLD reports have involved only descriptive statistical analyses. Herein we present the first multivariate risk factor analysis for variables that influence survival in KTX with PTLD.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

An analysis of all United States KTX voluntarily reported to the IPITTR with PTLD from November 1968 to January 2000 was retrospectively carried out in December 2001. Factors evaluated included age at transplant, gender, race, type of transplant (living or cadaveric), age at PTLD diagnosis, time of occurrence of PTLD, immunosuppression, PTLD characteristics (Epstein–Barr virus status, morphology, clonality), sites of PTLD presentation, treatment of PTLD, causes of death and patient survival.

Statistical analysis

A Cox regression model was performed to explain survival outcomes. A Kaplan–Meier time to event analysis was performed for survival (Tables 1 and 2). A p-value of less than 0.05 was considered to be statistically significant.

Table 1.  Multivariate analysis of risk factors for survival in patients diagnosed with PTLD
VariableSurvivalHazard ratio (HR)p-value
  1. ns = not significant. A p-value of less than 0.05 was considered to be statistically significant.

Sites
 Single vs. multiple47% vs. 27%1.4p < 0.04
B Cell (yes/no)49% vs. 33%1.5p < 0.03
ATG/ALG,OKT344% vs. 35%0.52p < 0.002
EBV (yes/no)54% vs. 50%1.5p = ns
CYA (yes/no)43% vs. 33%1.1p = ns
AZA (yes/no)42% vs. 39%1.5p = ns
Gender
 Female vs. male45% vs. 42%1.2p = ns
Race
 Caucasian vs. African American45% vs. 45%1.5p = ns
Age at diagnosisA 1-year increase in age increases the risk of death by 2%0.98p = 0.0004
Table 2.  Univariate analysis of risk factors for survival in patients diagnosed with PTLD
TreatmentSurvivalp-value
  1. *Note: 100% of patients with Nephrectomy had ISD. while only 30% of those without nephrectomy had ISR/ISD.

Surgery for PTLD (yes/no)55% vs. 0%p < 0.0001
Transplant nephrectomy for isolated allograft alone PTLD involvement80% vs. 53%p = ns
Patients diagnosed with PTLD < or > 6 months post-transplant36% vs. 46%p < 0.0013
Patients diagnosed with PTLD < or > 12 months post-transplant40% vs. 45%p < 0.04

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Review of the IPITTR data from November 1968 to January 2000 revealed 402 KTX who developed PTLD. Mean age at transplant was 42 ± 16.3 years and mean age at diagnosis was 43 ± 16.4 years, 36(8%) patients were under the age of 18 years at the time of diagnosis. Age at diagnosis was significantly associated with death risk, with a 1-unit increase in age increasing the death risk by 2% (HR 0.98, p < 0.0004) (Table 1). Kidney donor source consisted of 312 (78%) cadaveric renal allograft recipients, 87 (22%) living related donor kidney recipients and 3 (1%) living unrelated donor kidney recipients. Ethnicity distribution included 269 (67%) Caucasian, 44 (11%) African-American and 89 (22%) other patients. Gender distribution data consisted of 277 (69%) male and 125 (31%) female recipients. There was no significant difference in risk of death when comparing gender or race (Table 3).

Table 3.  Demographics
No. of kidney transplant PTLD patientsN = 402
Age at transplant42 ± 16.3 years
Kidney donor source
 Cadaveric312 (77.6%)
 Living related donor87 (21.6%)
 Living unrelated donor3 (0.8%)
Dates of transplantation1968–1998
Ethnic groups
 Caucasian269 (67%)
 African American44(11%)
 Other89 (22%)
Gender
 Male277 (69%)
 Female125 (31%)
Antilymphocyte therapy for induction/rejection
   Polyclonal preparations (ALG/ATG)173 (43%)
   Monoclonal preparation (OKT3)165 (41%)
   OKT3+ALG/ATG64 (16%)
Maintenance immunosuppression
 Cyclosporine, azathioprine, prednisone306 (76%)
 Tacrolimus, mycophenolate  mofetil, prednisone84 (21%)
 Other regimens (azathioprine, prednisone)12 (3%)

Immunosuppression

Maintenance immunosuppression consisted of cyclosporine, azathioprine and prednisone in 306 (76%) patients, tacrolimus, mycophenolate mofetil and prednisone in 84 (21%) patients and other regimens such as azathioprine and prednisone in 12 (3%) patients. Antibody preparations were administered for either induction or rejection treatment and consisted of polyclonal preparations (antithymocyte {ATG} or antilymphocyte {ALG} preparations) in 173 (43%) of the patients, murine monoclonal antibody (OKT3) in 165 (41%) of the patients and 64 (16%) of the patients received ATG/ALG and OKT3. (Table 3). There was a significant difference in the death risk when comparing patients who received ATG/ALG, OKT3 or no antibody therapy (HR 0.52, p < 0.002). There was no significant difference in death risk in patients who received cyclosporine versus those who received tacrolimus, or when comparing patients who received azathioprine versus those who received mycophenolate mofetil (Table 1).

PTLD characteristics and presentation

Median time to PTLD diagnosis was 18 (range 0–310) months post-transplant. However, when patients were analyzed by decade, in patients with histologic data available, 86% were B cell predominant, 64% monoclonal, 81% EBV positive and 85% polymorphic. There was a significant difference in death risk when comparing B cell predominant PTLDs to non-B cell predominant PTLDs (HR 1.48, p < 0.03). Interestingly though, there was no difference in death risk when comparing EBV status. Almost half the patients (49%) presented with two or more sites of PTLD involvement and 51% presented with a single site. Whereas lymphomas in the general population frequently involve lymph nodes, 68% of PTLDs occurred in extranodal sites, including the allograft (21%), central nervous system (19%), gastrointestinal tract (19%), liver (15%), lung (11%) and spleen (11%). Only 32% of patients presented with lymph node involvement (Table 4). A significant increase in death risk was observed with multiple (73%) versus single sites (53%) involvement of PTLD (HR 1.42, p < 0.04) (Table 1).

Table 4.  PTLD characteristics
Time to PTLD diagnosis post-transplant18 (0–310) months
PTLD histopathology
 B cell predominant86%
 Monoclonal64%
 EBV positive81%
 Polymorphic85%
PTLD presentation sites
 Kidney allograft alonen = 35 (9%)
 Kidney allograft plus other sitesn = 52 (13%)
 Other sites without kidney allograftn = 315 (78%)
Extrarenal sites of PTLD presentation  in patients with allograft involvement  plus other sitesn = 52
   Central nervous systemn = 17 (9%)
   Gastrointestinaln = 19 (10%)
   Spleenn = 21 (11%)
   Livern = 23 (12%)
   Lymph nodesn = 31 (16%)
Causes of death
 PTLDn = 234 (58%)
 Othern = 64 (16%)
 Cardiacn = 48 (12%)
 Infectionn = 56 (14%)

PTLD therapy

In the IPITTR series, 46 (11%) patients died without treatment, either because the diagnosis was missed or was made post-mortem. Of the 356 (89%) patients who received therapy for PTLD, monotherapy was utilized in 235 (66%) and consisted of immunosuppression discontinuation (ISD) (48%), surgical intervention (22%), immunosuppression reduction (ISR) (19%) or radiation (11%). Combination therapy was used in 121 (34%) of patients and consisted of ISR or ISD with chemotherapy (62%), nephrectomy (16%), surgery (12%) or radiation (10%). Survival by therapeutic approach included ISD 48%, ISR 64%, surgery 53% (allograft nephrectomy (n = 47), native nephrectomy (n = 3), bowel resection (n = 15), lobectomy (n = 2), orchiectomy (n = 1) parotidectomy (n = 1), craniotomy (n = 2), spleenectomy (n = 4), tonsillectomy (n = 4) and excision (n = 33)), ISR or ISD and radiation 51%, ISR or ISD and surgery 50%, ISR or ISD and chemotherapy 38%. No statistically significant differences were observed in survival rates according to the type of therapy.

Of the patients with PTLD isolated to the allograft (N = 35), 57% (n = 20) were treated with transplant nephrectomy alone and 80% (n = 16) survived, compared with 53% survival observed in patients for whom the allograft was not removed (n = 15), however, these differences did not reach statistical significance. Survival may have been influenced by the difference in immunosuppression, however, 100% of the patients who had a transplant nephrectomy had ISD, while only 30% of patients who did not undergo a transplant nephrectomy had ISR or ISD. Overall survival in patients with PTLD involvement of the allograft alone was 69%. In contrast, patients with other organ involvement in addition to the allograft, survival was only 36% (n = 19, p < 0.0001). By univariate analysis, there was no difference in survival between patients with kidney allograft involvement and other sites versus other site involvement without the allograft (p = 0.1963). Patients who received surgical therapy had higher survival (55%) compared to those who were not treated surgically (0%) (p < 0.0001) (Table 2, Figure 1).

image

Figure 1. Kaplan–Meier survival analysis for renal transplant recipients with PTLD.

Download figure to PowerPoint

Survival data

The death rate in patients who were diagnosed with PTLD within 6 months post-transplant (64%) was greater than that in patients diagnosed beyond 6 months post-transplant (54%, p < 0.0013). Differences were also observed in death risks between patients diagnosed with PTLD less than 1 year post-transplant (60%) compared to those who were diagnosed with PTLD greater than 1 year post-transplant (55%, p < 0.04) (Table 2).

In addition to an approximate 30% reduction in patient survival within the first year following PTLD diagnosis, PTLD patients continue to have an ongoing increased risk of death following the first year after PTLD diagnosis. The reason for this continued increase in death risk is not clear from the present experience, however, 58% (n = 233) of the patients died of PTLD related complications at a median of 2 (range 0–145) months, 16% died of other causes at a median of 21.2 (range 0–145) months, 12% died of cardiac complications at a median of 7 (range 0–145) months, 7% died of infection at a median of 3 (range 0–124) months and 14% died of infection at a median of 3.5 (0–124) months. Of patients who died, only 9 (3%) did not have multiple site involvement at the time of death. Only in two patients was the death related to kidney transplant graft loss (Table 4).

Analysis by decade

Patients affected with PTLD were also analyzed by decade of transplant. Two (0.5%) patients transplanted during 1960–1969 developed PTLD at a median of 23.9 years post-transplant and both (100%) subsequently died. Fourteen (3.5%) patients transplanted during 1970–1979 developed PTLD at a median of 12.4 years post-transplant and 43% of the patients died. Due to the small number of patients in these two decades, it is difficult to ascertain risk factors for survival. Interestingly, the patients transplanted in the next two decades (1980–1989 and 1990–1999) developed PTLD more quickly than the previous two decades. One hundred fifty-seven (39%) patients transplanted during 1980–1989 developed PTLD at a median of 3.2 years post-transplant and 229 (57%) of patients transplanted during 1990–1999 developed PTLD at a median of 0.7 years post-transplant. Sixty-eight percent of PTLD patients transplanted during 1980–1989 and 51% of patients transplanted during 1990–1999 died. However, survival did not differ greatly in these groups compared to the previous two decades. A separate risk factor survival analysis of patients who developed PTLD in the 1980s versus those who developed PTLD in the 1990s did not reveal any differences in overall survival or in terms of the effects of any risk factor from the survival analysis.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

The present series represents a multivariate risk factor analysis performed on the largest experience of PTLD in KTX reported to date. Time from transplantation to PTLD presentation varies widely in the literature, ranging from less than 1 month to several years. In this series, PTLD presented relatively early post-transplant, occurring at a median of 18 months.

The incidence of PTLD in pediatric transplant recipients has been reported as higher than that in adults, and is thought to be related to a higher rate of EBV infection in children (10,11). Interestingly, our multivariate analysis revealed that, at time of PTLD diagnosis, for each year of increased age the risk of death rose by 2%. This observation suggests a bimodal effect of age on PTLD risk. No difference in survival between patients of different gender or race was determined (Table 1).

In this IPITTR experience, the majority of patients with PTLD were on cyclosporine, azathioprine, prednisone based regimens. This observation, however, may reflect the historic nature of IPITTR as PTLD has been noted in all transplant immunosuppressive eras. A significant difference in survival was observed between patients who received ATG/ALG or OKT3 therapy for induction or rejection versus those who did not. This may be reflective of the risks associated with greater amounts of immunosuppression exposure in these patients.

Cancer development is linked to the immunosuppressed state per se which among other factors, makes transplant recipients prone to a variety of virus-related tumors (10,12–14). Transplant recipients, have demonstrated a susceptibility to infection from viruses such as EBV and cytomegalovirus which have been implicated as co-factors in malignancy formation (12,15–18). The majority of the PTLDs in this series were B cell predominant, EBV positive, monoclonal and polymorphic. Our analysis did find B cell predominance as a factor associated with lower survival. Interestingly, EBV serologic status did not influence survival. However, this may be reflective of the limited amount of data available for this variable.

In distinct contrast to lymphomas in the general population where lymph nodes are almost always involved, 84% of PTLDs did not involve lymph nodes. Surprisingly, one of the most frequent extra-nodal sites was the central nervous system, which was involved in 19% of patients with PTLD. The reasons for neurotropism, however, are not clear from the present series.

Allograft involvement represents an interesting feature of PTLD. A relatively high proportion (21%) of patients presented with renal allograft involvement. Patients with PTLD isolated to the allograft may be surgically cured and often undergo ISD, possibly allowing reconstitution of the immune system to act against PTLD and more adequately prevent deadly opportunistic infections. In the IPITTR series, survival in patients with PTLD restricted to the renal allograft alone who underwent transplant nephrectomy was increased compared to those patients who did not have an allograft nephrectomy, although this was not statistically significant. This observation is similar to the more general observation where patients with single site involvement of PTLD had superior survival compared to those with multiple site involvement. This may be due to the greater extent of disease, which requires increased intensity of therapy and therefore greater toxicities. It is not possible to determine from the present experience whether the increased survival in patients with isolated involvement of the renal allograft is a function of (1) a fundamentally different biologic behavior of PTLD in these patients, (2) allograft removal, (3) the degree of ISR or (4) a combination of all of these factors.

The death rate in patients diagnosed with PTLD within 6 months post-transplant (64%) was greater than that in patients diagnosed beyond 6 months post-transplant (54%, p = 0.04). Similarly, 60% of patients within 12 months from transplant died compared to 55% of those after 1-year post-transplant (p < 0.04). There are several possible explanations for this including (1) late PTLD recurrence, (2) toxicities from PTLD therapy and (3) late presentation of opportunistic infections.

Our analysis also demonstrated that death risk increased with increasing age at diagnosis. The reasons for this finding are not apparent from the present study. However, older patients may not tolerate chemotherapy as well as younger patients due to concomitant medical complications such as cardiovascular disease, and toxicities in this patient population are not as well known, as elderly patients are often underrepresented in oncology therapeutic intervention trials (19,20).

This first multivariate analysis of IPITTR data provides insight into various questions regarding PTLD in KTX. Although the time of PTLD diagnosis decreased from a median of 23.9 years post-transplant in the 1960s to a median of 0.7 years post-transplant in the 1990s, no difference in survival was observed between 1960 and 1999. Moreover, when comparing the larger numbers of patients diagnosed with PTLD in 1980 to those diagnosed with PTLD in 1990, no significant difference was observed in terms of the effects of any risk factors on survival. We hypothesize that the decrease in the median time to PTLD diagnosis is the result of the development of more potent immunosuppression and the transplantation of higher immunologic risk patients that require greater amounts of immunosuppression for the prevention of rejection. Improvements in treatment for PTLD may also account for the lack of difference in survival throughout the decades, despite an earlier occurrence of PTLD in the 1990s. Future research should be focused on defining risk factors for PTLD in the current immunosuppressive era, and further analyzing outcomes in patients with PTLD.

Overall, risk factors that influence survival demonstrated a significantly higher death risk with multiple site PTLD involvement, B cell predominance, exposure to ALG/ATG, OKT3 and increasing age. Univariate analysis also found increased death risk for renal transplant recipients who did not undergo surgery. Additionally, survival in patients with PTLD restricted to the renal allograft alone who underwent transplant nephrectomy was higher than that in those patients who did not have an allograft nephrectomy, although this did not reach statistical significance. The death rate in patients diagnosed with PTLD within 6 and 12 months post-transplant was also found to be greater than that in patients diagnosed beyond 12 months post-transplant and although further research is needed, we hypothesize that this may be related to a more aggressive PTLD. A limitation of this study is that due to the voluntary nature of the registry, we are not able to analyze the underlying transplant population from which these PTLD cases were derived. Future studies are warranted for further comparison of risk factors and outcomes.

Currently, there is no uniform standard for defining PTLD, and PTLD treatment varies widely. Due in part to the small number of patients affected by PTLD at various transplant centers, there are no randomized-controlled trials to determine the efficacy of preventative or therapeutic interventions. Efforts should also focus on the prevention of PTLD. Current and future research efforts should be aimed at (1) identifying risk factors for PTLD, (2) standardizing a PTLD definition and (3) evaluating the role of therapeutic interventions to improved patient survival in renal transplant recipients affected with PTLD.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References
  • 1
    Penn I, Hammond W, Brettschneider L, Starzl TE. Malignant lymphomas in transplantation patients. Transplant Proc 1969; 1: 106112.
  • 2
    Trofe J, Buell JF, First MR, Hanaway MJ, Beebe TM, Woodle ES. The role of immunosuppression in lymphoma. Recent Results Cancer Res 2002; 159: 5566.
  • 3
    Birkeland SA, Hamilton-Dutoit S. Is posttransplant lymphoproliferative disorder (PTLD) caused by any specific immunosuppressive drug or by the transplantation per se? Transplantation 2003; 76: 984988.
  • 4
    Funch DP, Brady J, Ko HH, Dreyer NA, Walker AM. Methods and objectives of a large US multicenter case-control study of post-transplant lymphoproliferative disorder in renal transplant patients. Recent Results Cancer Res 2002; 159: 8188.
  • 5
    Libertiny G, Watson CJ, Gray DW, Welsh KI, Morris PJ. Rising incidence of post-transplant lymphoproliferative disease in kidney transplant recipients. Br J Surg 2001; 88: 13301334.DOI: 10.1046/j.0007-1323.2001.01924.x
  • 6
    Loren AW, Porter DL, Stadtmauer EA, Tsai DE. Post-transplant lymphoproliferative disorder: a review. Bone Marrow Transplant 2003; 31: 145155.DOI: 10.1038/sj.bmt.1703806
  • 7
    Leblond V, Dhedin N, Mamzer Bruneel MF, Choquet S, Hermine O, Porcher R et al. Identification of prognostic factors in 61 patients with posttransplantation lymphoproliferative disorders. J Clin Oncol 2001; 19: 772778.
  • 8
    Tsai DE, Hardy CL, Tomaszewski JE, Kotloff RM, Oltoff KM, Somer BG et al. Reduction in immunosuppression as initial therapy for posttransplant lymphoproliferative disorder: analysis of prognostic variables and long-term follow-up of 42 adult patients. Transplantation 2001; 71: 10761088.DOI: 10.1097/00007890-200104270-00012
  • 9
    Choquet S, Mamzer BM, Hermine O, Porcher R, Nguyen QS, Davi F et al. Identification of prognostic factors in post-transplant lymphoproliferative disorders. Recent Results Cancer Res 2002; 159: 6780.
  • 10
    Penn I. Cancers in renal transplant recipients. Adv Ren Replace Ther 2000; 7: 147156.
  • 11
    Shapiro R, Nalesnik M, McCauley J, Fedorek S, Jordan ML, Scantlebury VP et al. Posttransplant lymphoproliferative disorders in adult and pediatric renal transplant patients receiving tacrolimus-based immunosuppression. Transplantation 1999; 68: 18511854.DOI: 10.1097/00007890-199912270-00006
  • 12
    Shiel A. Cancer in Dialysis and Transplant Patients. In: MorrisPJ, ed. Kidney Transplantation Principles and Practice. 5th Ed. Philadelphia : WB Saunders; 2001. p. 558570.
  • 13
    Preiksaitis JK. Epstein–Barr virus infection and malignancy in solid organ transplant recipients: strategies for prevention and treatment. Transpl Infect Dis 2001; 3: 5659.DOI: 10.1034/j.1399-3062.2001.003002056.x
  • 14
    Green M, Reyes J, Webber S, Rowe D. The role of antiviral and immunoglobulin therapy in the prevention of Epstein–Barr virus infection and post-transplant lymphoproliferative disease following solid organ transplantation. Transpl Infect Dis 2001; 3: 97103.DOI: 10.1034/j.1399-3062.2001.003002097.x
  • 15
    Holmes RD, Sokol RJ. Epstein–Barr virus and post-transplant lymphoproliferative disease. Pediatr Transplant 2002; 6: 456464.DOI: 10.1034/j.1399-3046.2002.02043.x
  • 16
    Cosio FG, Nuovo M, Delgado L, Yearsley M, Porcu P, Caligiuri M et al. EBV kidney allograft infection: possible relationship with a peri-graft localization of PTLD. Am J Transplant 2004; 4: 116123.
  • 17
    Ho M, Jaffe R, Miller G, Breinig MK, Dummer JS, Makowka L et al. The frequency of Epstein–Barr virus infection and associated lymphoproliferative syndrome after transplantation and its manifestations in children. Transplantation 1988; 45: 719727.
  • 18
    Hanto DW, Frizzera G, Gajl-Peczalska J, Purtilo DT, Klein G, Simmons RL et al. The Epstein–Barr virus (EBV) in the pathogenesis of posttransplant lymphoma. Transplant Proc 1981; 13: 756760.
  • 19
    Argiris A, Li Y, Murphy BA, Langer CJ, Forastiere AA. Outcome of elderly patients with recurrent or metastatic head and neck cancer treated with cisplatin-based chemotherapy. J Clin Oncol 2004; 22: 262268.DOI: 10.1200/JCO.2004.08.039
  • 20
    Au HJ, Mulder KE, Fields AL. Systematic review of management of colorectal cancer in elderly patients. Clin Colorectal Cancer 2003; 3: 165171.