Although long-term survival after solid organ transplantation is continuing to increase, recipients remain at increased risk of premature death,1-5 primarily because of an increased mortality from cardiovascular disease, cancer, infection, and recurrent disease. Cytomegalovirus (CMV) infection is common, affecting 50% to 80% of the adult population with an initial, often subclinical mild infection followed by a state of lifelong chronic viral carriage.6 The CMV infection is generally well controlled in an immune competent host, but in the immunosuppressed, it is often associated with significant morbidity. Whereas some studies both in the normal and the transplant population have found an increased risk of cardiovascular disease and death in CMV infected individuals,4,7-9 this has not been confirmed by all.10-13 However, all the studies in the transplant population have been relatively short term (less than 5 years).
Some viruses, such as human papilloma virus, hepatitis B and C, Epstein-Barr virus, and human herpes virus-8 have an established role in the development of cancer both in the immunocompetent and the immunosuppressed organ allograft recipient populations. Although CMV antigens have been identified in cells of certain tumors, such as cancers of the colon, prostate, lymphoma, and glioblastoma,14 it is not known if the presence of CMV is an epiphenomenon or whether there is a causative association. The role of CMV in posttransplantation cancer is also controversial with conflicting evidence from recent studies: some have shown a protective effect,15 whereas others, an increased risk.16
We therefore examined a large national cohort of solid allograft recipients for the impact of CMV in donors and recipients on the long-term posttransplantation survival. We have also investigated for an association between CMV and the risk of posttransplantation cancer.
Using the data held by the U.K. Transplant Registry, the recipients of first solid organ (kidney, liver heart, and lung) transplantation between January 01, 1987, and December 31, 2007 who were resident in England, Wales or Scotland were identified. The transplants where the CMV status was not recorded for the donor or the recipient were excluded (n = 12228). Recipients of combined kidney-pancreas (n = 764) were grouped with the kidney recipients and recipients of double-lung or heart-lung transplants (n = 553) were grouped with the lung recipients. The recipients were divided into 4 groups based on the combination of donor and recipient CMV immunoglobulin G status at the time of transplantation: both donor and recipient CMV positive (D+ R+) or negative (D− R−), CMV-positive donor with CMV-negative recipient (D+ R−) and CMV-negative donor with CMV-positive recipient (D− R+).
Transplant and CMV data were obtained from the U.K. Transplant Registry. Cancer registration data were obtained from the Office for National Statistics by matching the details of the recipients (name, sex, date of birth/death, address, and National Health Service [NHS] number). All types of cancers other than non-melanoma skin cancer were included. The data for the first diagnosed cancer after transplantation were included. To ignore the cancers which may have been present at the time of transplantation, all cancers other than Hodgkin lymphoma or non-Hodgkin lymphoma diagnosed within a month of transplantation were excluded.
Kaplan-Meier analysis and log-rank test were used to assess the long-term posttransplantation patient survival. The risk-adjusted hazard of death was calculated using Cox proportional hazards model, correcting for those risk factors that have been identified by NHS Blood and Transplant as having a significant impact on survival.17 Risk adjusted hazard of death for different solid organ recipients are shown separately because the risk adjustment varies. Risk factors included were: donor age, recipient age, donor sex, recipient sex, donor type (kidney and lung recipients), donor cause of death (kidney and heart recipients), transplant year, primary disease (kidney, liver and heart recipients), HLA mismatch (kidney recipients), and ischemia time (cold ischemia for liver, total for heart/lung). For this assessment, death of the recipient within 10 years (or 1 year, when 1-year survival was assessed) was considered as an event, and the recipients who were alive at 10 years (or 1 year, when 1-year survival was assessed) from transplantation and those who were lost for follow-up were censored. During Cox regression, missing data for a categorical covariate were grouped together as “unknown” category. Recipients with missing ischemia time (12% of liver recipients, 35% of cardiothoracic recipients) were excluded from multivariate analysis using Cox regression only. Times to diagnosis of cancer in the 4 CMV groups were compared using the log-rank test. Cox proportional hazards modelling was also used to assess the relationship between various CMV groups and diagnosis of cancer within 10 years of transplantation. The factors used for risk adjustment in this analysis were recipient age and sex. For this assessment, a diagnosis of cancer within 10 years of transplantation was considered as an event and any other end point in the absence of a diagnosis of cancer, such as death or end of follow-up, was considered as censored. While comparing the hazard of multiple types of cancers in different CMV groups, the Bonferroni correction was used to assess statistical significance to avoid inflating the overall type I error above its nominal level for each test. The causes of death between the D− R− groups and other recipient group were compared using the χ2 test. All data were analyzed using SAS version 9.1.3 (SAS Institute, Cary).
A total of 22461 recipients were studied, including 13,215 (59%) kidney recipients, 4814 (21%) liver recipients, 2686 (12%) heart recipients, and 1746 (8%) lung recipients. The baseline characteristics of the recipients in different CMV groups are shown in Table 1.
Survival curves for the recipients of all organ transplantation over 10 years from transplantation are shown in Figure 1. At 10 years from transplantation, survival of D− R− group (73.6 [95%CI, 72.3, 74.9]) was significantly higher (P < 0.0001) compared with the combined survival of all the other recipients (66.1% [65.3, 66.9]). The donor-recipient CMV matching was associated with significant survival advantage for CMV-negative recipients (10-year survival: 73.6% [72.3, 74.9] for D− R− group and 68.4% [66.9, 69.9] for D+ R− group) but not for CMV-positive recipients (10-year survival: 64.6% [63.3, 65.8] for D+ R+ group and 66.1% [64.7, 67.5] for D− R+ group). Table 2 shows the risk-adjusted hazard of death within 1 year and 10 years from transplantation, assessed separately for the recipients of kidney, liver, heart, and lung transplantation. A total of 6213 recipients died within 10 years from transplantation. The recipients who died within 30 days of transplantation (n = 897) were excluded because CMV is not likely to have contributed to death in this group. A majority of these (471, 53%) died within 8 days of transplantation, and the cause of death in this group is more likely to be related to perioperative complications than CMV. The causes of death were studied in the remaining 5316 recipients after dividing them into D− R− group and “others” group, and the results are shown in Figure 2.
The unadjusted incidence of all cancers was 8.8% (7.9, 9.6) among D+ R+ group, 7.0% (6.0, 7.9) among D+ R− group, 9.1% (8.1, 10.1) among D− R+ group, and 6.4% (5.5, 7.3) among D− R− group, and this difference was statistically significant (P < 0 · 0001). However, there was no statistically significant difference in the risk-adjusted hazard of all cancers between these groups after correction for age and sex (Figure 3). Similar results were obtained on comparing risk of cancer among D−R− group against the risk of cancer among all other recipients. The unadjusted hazard of cancer among all other recipients was higher compared to D−R− recipients (hazard ratio [HR] of 1.37 among other recipients compared to D−R− group, P < 0.0001) but after adjustment for age and sex, the difference in hazard of cancer between the 2 groups was not statistically significant (HR of 1.036 among other recipients, P = 0.63). In our cohort, along with adjustment of risk for age (which is a recognized risk factor for cancer), we also adjusted for sex because male sex was found to be an independent risk factor for cancer even after excluding those cancers which are sex-specific, such as cancer of cervix, ovary uterus, and prostate (HR for male sex, 1.36; P < 0.0001).
The impact of CMV status was studied on the risk of kidney cancer among kidney recipients: there were 72 cases of kidney cancers among 13,215 kidney recipients. Compared with D−R− recipients, other recipients had a higher unadjusted hazard of developing kidney cancer after kidney transplantation (HR, 2.33; P = 0.036) but after adjustment for age and sex, the difference in the hazard was not statistically significant (risk adjusted HR, 1.98; P = 0.096).
(Table S1, SDC, http://links.lww.com/TP/B123) shows the frequency and unadjusted incidence of 23 types of cancers among the recipients in different CMV groups. The risk-adjusted hazard of developing different types of cancers is shown as forest plots in (Figure S1, SDC, http://links.lww.com/TP/B123).
This study found that the use of organs from donors who are CMV-positive for recipients who are CMV-negative is associated with an increased long-term posttransplantation mortality in kidney, heart, and lung transplant recipients.
The effect of CMV on posttransplantation patient survival has been studied among recipients of different organs with conflicting results. The published reports assessing patient survival after transplantation include studies of kidney recipients11 (no effect), kidney-pancreas recipients12 (no effect), liver recipients13 (no effect), and heart recipients9 (7.05 times increased hazard of mortality at 1 year). Most of these studies have assessed short-term survival, often limited to less than 5 years from transplantation. The data assessing the long-term survival are limited. One of the larger reports is the study by Johnson and colleagues,11 which included 8228 recipients of deceased donor kidney transplantation in the United Kingdom between 2000 and 2007, partly overlapping with our present cohort. Johnson demonstrated no effect of CMV on the posttransplantation patient survival at 1, 3, and 5 years. In comparison, our study showed an increased risk of patient death at 10 years after transplantation among recipients of kidney as well as heart and lung transplantations. This difference is likely to be due to larger numbers and longer follow-up. A direct comparison between our cohort and the cohort studied by Johnson could not be performed due to inherent differences between the 2 cohorts, such as different inclusion criteria and unspecified duration of follow-up. In our study, an increased mortality was observed in the D+R− group at 10 years (14% increase in risk of death in comparison to D−R− group among kidney recipients, 34% increase among heart recipients and 35% increase among lung recipients) possibly highlighting the effect of the “new” CMV infection acquired during transplantation. An increased mortality was also observed among D+R+ group of heart recipients (31% increase) and lung recipients (27% increase).
It is difficult to explain the processes by which CMV may be contributing to increased posttransplantation mortality in a retrospective registry study, such as ours. The most common causes of death within 10 years after transplantation included cardiovascular events, cerebrovascular events, infections, and single or multiorgan failure. The analysis of causes of death showed a small (2%) increase in cardiovascular death among CMV-infected recipients; however, it should also be noted that causes of death in more than a third of recipients were unspecified or “others.” The CMV infection has been shown to be associated in the nontransplantation population with an increased cardiovascular mortality.4,5,8 Among solid organ transplantation recipients, CMV has been shown to be associated with a variety of conditions, such as acute rejection,18 tubulointerstitial nephritis and glomerulopathy after kidney transplantation,19 hepatic artery thrombosis and accelerated hepatitis C virus infection20 after liver transplantation, allograft vasculopathy after heart transplantation and bronchiolitis obliterans after lung transplantation,7 bacterial, fungal, and viral infections,21,22 and new onset diabetes mellitus.23 The rates of these complications and their impact on posttransplantation mortality could not be assessed in the present cohort due to lack of relevant data in the retrospective audit. The increased mortality observed among CMV-infected transplantation recipients in our cohort may be interplay of some or all of these diseases.
In the general population, especially the healthy elderly population, CMV is described as a driver of age-associated immune alterations leading to a reduction in naive T cells.24 Reactivation of CMV may result in increased levels of proinflammatory cytokines, such as interleukin-6 and tumour necrosis factor-α. C-reactive protein levels also increase as a consequence of leakage of the virus from host cells via the action of interleukin-6. These inflammatory markers have been linked to both all-cause and cardiovascular disease–related mortality. Recently, Savva et al25 found in a cohort of 511 healthy individuals aged at least 65 years followed up for 18 years that CMV infection was associated with an increased mortality rate and a near doubling cardiovascular death, whereas there was no increase in mortality from other causes. Simanek et al,26 in a large and younger representative American population, aged 25 years or older with up to 18 years of follow-up, showed that CMV seropositivity was independently associated with an increased all-cause mortality (adjustment for CRP level did not attenuate this relationship). However, after confounder adjustment, they failed to associate CMV serostatus with cardiovascular mortality. Another study by Courivaud et al27 showed that CMV exposure was an independent risk factor for atherosclerotic events, whereas posttransplantation CMV replication was an independent risk factor for both atherosclerotic events and death in kidney transplant recipients. Their results suggested that CMV was associated with immune exhaustion and inflammation in favor of an indirect effect of CMV on atherosclerotic progression. These results, along with the results of our study, provide evidence toward a complex interaction between aging, CMV infection, and cardiovascular disease/death both in general population as well as recipients of organ transplantation.
CMV Does Not Have an Effect on the Risk of Posttransplantation Cancer
The cancer risk of a transplantation recipient is influenced by complex interactions between recipient factors (age, ethnicity, social status, premalignant conditions, pre-existing infections, smoking, alcohol intake, possibly genetic factors), donor factors (diseases transmitted from donor organ), and posttransplant factors (medical follow-up, long-term immunosuppression, drug toxicity, de novo infections). The recipient factors, which are associated with an increased risk of cancer, such as age, social status, and smoking, are also associated with higher CMV seroprevalence,28 and it can be difficult to tease out any independent effect exerted by CMV. A higher risk of cancer among CMV-positive recipients as compared to D−R− recipients observed on univariate analysis of our cohort may also be explained by the fact that CMV-positive recipients were older than CMV-negative recipients. This difference in the risk was not statistically significant after correction for recipient's age and sex.
From this large cohort of solid organ recipients, we demonstrated that the donor-recipient CMV statuses have no independent association with the risk of posttransplantation cancer in general or with the specific risk of developing 23 types of posttransplantation cancers. On univariate analysis, the incidence of posttransplantation cancer was higher among the groups with a CMV-positive recipient (D+R+ and D−R+) as compared to the groups with a CMV-negative recipient (D+R− and D−R−), but this difference was not statistically significant on multivariate analysis after correction for age and sex (Figure 3). This result highlights the fact that age is a risk factor for both CMV positivity and cancer.
The size of the cohort is a clear strength of our study and is likely to be the reason for the differences in findings between our study and the 2 other recent reports assessing a similar question.15,16 Couzi and colleagues15 retrospectively studied a cohort of 105 kidney recipients followed up for a median of 5 years, 23 of whom had developed a posttransplantation cancer. They concluded that CMV naive recipients had a 5.28 times increased risk of posttransplant cancer as compared to recipients exposed to CMV before or after transplantation and attributed this to CMV-mediated increase in the number of a subset of γδ T lymphocytes which possess antitumor activity. In contrast, Courivaud and colleagues16 reported from a cohort of 455 recipients of kidney transplantation that both pretransplantation exposure (HR = 1.8) and posttransplantation replication (HR = 2.17) of CMV were associated with an increased risk of posttransplantation cancer as compared to CMV naive recipients and attributed this to immune exhaustion related to exposure to CMV. Compared to these studies, our cohort was substantially larger resulting in increased statistical power to detect any association between CMV and posttransplantation cancer. As shown in (Table S1, SDC, http://links.lww.com/TP/B123), despite the cohort size, the number of individual types of cancers is relatively small; so studies with a smaller cohort would have very small numbers of individual types of cancer and are likely to produce results which cannot be generalized to all transplantation recipients. Despite the limitations of our study, as listed below, we believe that the validity of our results is enhanced by the large cohort size.
Limitations of our Study
As with all retrospective registry studies, our study has several important limitations. The CMV data were not available for 12,228 (35%) recipients, and these were excluded. The CMV status was recorded at the time of transplantation, and any recipients with posttransplantation acquisition of de novo CMV infection were not identified. Furthermore, in a significant proportion of patients, no case of death was specified. The data regarding pharmacological prophylaxis against CMV infection was not available. The risk of cancer would be underestimated in cases where the recipient had multiple cancers because the data for the first diagnosed posttransplantation cancer only was included. The cause of death data was obtained from the Office for National Statistics, which is the United Kingdom's largest independent producer of official statistics and is the most widely recognized national statistical institute in the United Kingdom. Multiple robust internal and external quality control measures are in place to maintain the high quality of data produced by the Office for National Statistics. Despite of this, some inaccuracy of data may be inevitable.
In summary, in this large cohort of solid organ recipients, there was no association between the donor-recipient CMV status and the risk of posttransplantation cancer as a whole, or the risk of 23 individual types of cancers. Novel findings also include the negative impact of the D+R− CMV mismatch on the long-term survival of recipients of kidney, lung, or heart transplantations but not the recipients of liver transplantation.
The authors are grateful to NHS Blood and Transplant for funding and the data relating to the donors and the recipients, and to the Office for National Statistics for providing the cancer data.
1. Lindholm A, Albrechtsen D, Frodin L, Tufveson G, Persson NH, Lundgren G. Ischemic heart disease—major cause of death and graft loss after renal transplantation in Scandinavia. Transplantation
. 1995; 60: 451–457.
2. Morales JM, Marcen R, del Castillo D, et al. Risk factors for graft loss and mortality after renal transplantation according to recipient age: a prospective multicentre study. Nephrol Dial Transplant
. 2012; 27(Suppl 4): iv39–iv46.
3. Pruthi J, Medkiff KA, Esrason KT, et al. Analysis of causes of death in liver transplant recipients who survived more than 3 years. Liver Transplant
. 2001; 7: 811–815. PubMed PMID: 11552217.
4. Blankenberg S, Rupprecht HJ, Bickel C, et al. Cytomegalovirus infection with interleukin-6 response predicts cardiac mortality in patients with coronary artery disease. Circulation
. 2001; 103: 2915–2921.
5. Muhlestein JB, Horne BD, Carlquist JF, et al. Cytomegalovirus seropositivity and C-reactive protein have independent and combined predictive value for mortality in patients with angiographically demonstrated coronary artery disease. Circulation
. 2000; 102: 1917–1923.
6. Centers for Disease Control and Prevention: Cytomegalovirus (CMV) and congenital CMV infection: Centers for Disease Control and Prevention; 2010 [cited 2013 18 August]. Available from: http://www.cdc.gov/cmv/overview.html
7. Thomas LD, Milstone AP, Miller GG, Loyd JE, Stephen Dummer J. Long-term outcomes of cytomegalovirus infection and disease after lung or heart-lung transplantation with a delayed ganciclovir regimen. Clin Transplant
. 2009; 23: 476–483.
8. Zhou YF, Leon MB, Waclawiw MA, et al. Association between prior cytomegalovirus infection and the risk of restenosis after coronary atherectomy. N Engl J Med
. 1996; 335: 624–630.
9. Li F, Kenyon KW, Kirby KA, Fishbein DP, Boeckh M, Limaye AP. Incidence and clinical features of ganciclovir-resistant cytomegalovirus disease in heart transplant recipients. Clin Infect Dis
. 2007; 45: 439–447. PubMed PMID: 17638191.
10. Zhu J, Quyyumi AA, Norman JE, Csako G, Epstein SE. Cytomegalovirus in the pathogenesis of atherosclerosis: the role of inflammation as reflected by elevated C-reactive protein levels. J Am Coll Cardiol
. 1999; 34: 1738–1743.
11. Johnson RJ, Clatworthy MR, Birch R, Hammad A, Bradley JA. CMV mismatch does not affect patient and graft survival in UK renal transplant recipients. Transplantation
. 2009; 88: 77–82.
12. Kaufman DB, Leventhal JR, Gallon LG, et al. Risk factors and impact of cytomegalovirus disease in simultaneous pancreas-kidney transplantation. Transplantation
. 2001; 72: 1940–1945.
13. Arthurs SK, Eid AJ, Pedersen RA, et al. Delayed-onset primary cytomegalovirus disease after liver transplantation. Liver Transplant
. 2007; 13: 1703–1709.
14. Soderberg-Naucler C. Does cytomegalovirus play a causative role in the development of various inflammatory diseases and cancer? J Intern Med
. 2006; 259: 219–246.
15. Couzi L, Levaillant Y, Jamai A, et al. Cytomegalovirus-induced gammadelta T cells associate with reduced cancer risk after kidney transplantation. J Am Soc Nephrol
. 2010; 21: 181–188.
16. Courivaud C, Bamoulid J, Gaugler B, et al. Cytomegalovirus exposure, immune exhaustion and cancer occurrence in renal transplant recipients. Transpl Int
. 2012; 25: 948–955. PubMed PMID: 22784509.
18. Nett PC, Heisey DM, Fernandez LA, Sollinger HW, Pirsch JD. Association of cytomegalovirus disease and acute rejection with graft loss in kidney transplantation. Transplantation
. 2004; 78: 1036–1041.
19. Rane S, Nada R, Minz M, Sakhuja V, Joshi K. Spectrum of cytomegalovirus-induced renal pathology in renal allograft recipients. Transplant Proc
. 2012; 44: 713–716.
20. Bosch W, Heckman MG, Pungpapong S, Diehl NN, Shalev JA, Hellinger WC. Association of cytomegalovirus infection and disease with recurrent hepatitis C after liver transplantation. Transplantation
. 2012; 93: 723–728.
21. George MJ, Snydman DR, Werner BG, et al. The independent role of cytomegalovirus as a risk factor for invasive fungal disease in orthotopic liver transplant recipients. Boston Center for Liver Transplantation CMVIG-Study Group. Cytogam, MedImmune, Inc. Gaithersburg, Maryland. Am J Med
. 1997; 103: 106–113.
22. Arthurs SK, Eid AJ, Pedersen RA, et al. Delayed-onset primary cytomegalovirus disease and the risk of allograft failure and mortality after kidney transplantation. Clin Infect Dis
. 2008; 46: 840–846.
23. Leung Ki EL, Venetz JP, Meylan P, Lamoth F, Ruiz J, Pascual M. Cytomegalovirus infection and new-onset post-transplant diabetes mellitus. Clin Transplant
. 2008; 22: 245–249.
24. Sansoni P, Vescovini R, Fagnoni FF, et al. New advances in CMV and immunosenescence. Exp Gerontol
. 2014; 55: 54–62.
25. Savva GM, Pachnio A, Kaul B, et al. Cytomegalovirus infection is associated with increased mortality in the older population. Aging Cell
. 2013; 12: 381–387.
26. Simanek AM, Dowd JB, Pawelec G, Melzer D, Dutta A, Aiello AE. Seropositivity to cytomegalovirus, inflammation, all-cause and cardiovascular disease-related mortality in the United States. PLoS One
. 2011; 6: e16103.
27. Courivaud C, Bamoulid J, Chalopin JM, et al. Cytomegalovirus exposure and cardiovascular disease in kidney transplant recipients. J Infect Dis
. 2013; 207: 1569–1575.
28. Dowd JB, Aiello AE, Alley DE. Socioeconomic disparities in the seroprevalence of cytomegalovirus infection in the US population: NHANES III. Epidemiol Infect
. 2009; 137: 58–65.