Biomarker And Therapy Intervention For Malignancy Risk Patients

Abstract

The invention relates to a method of determining an increased risk of cancer in an immunosuppressed patient, the method comprising: (a) determining the percentage of CD8+CD57+ T-cells in a population of CD8+ T-cells in a sample from the patient; wherein a percentage of 40% or greater of CD8+CD57+ T-cells is indicative of an increased risk of cancer; and/or (b) determining the percentage of CD4+CD57+ T-cells in a population of CD4+ T-cells in a sample from the patient; wherein a percentage of 10% or greater of CD4+CD57+ T-cells is indicative of an increased risk of cancer.

Claims

1. A method of determining an increased risk of cancer in an immunosuppressed patient, the method comprising: (a) determining the percentage of CD8+CD57+ T-cells in a population of CD8+ T-cells in a sample from the patient; wherein a percentage of 40% or greater of CD8+CD57+ T-cells is indicative of an increased risk of cancer; and/or (b) determining the percentage of CD4+CD57+ T-cells in a population of CD4+ T-cells in a sample from the patient; wherein a percentage of 10% or greater of CD4+CD57+ T-cells is indicative of an increased risk of cancer.

2. The method according to claim 1, wherein a percentage of 50% or greater of CD8+CD57+ T-cells is indicative of an increased risk of cancer.

3. The method according to any preceding claim, further comprising the step of selecting patients determined to be at increased risk of cancer for one or more of: increased surveillance for cancer; modifying the immunosuppressive therapy regime; or providing preventative therapy for cancer, and optionally wherein the preventative therapy for cancer is anti-cancer therapy.

4. The method according to any preceding claim, further comprising determining the percentage of CD8+CD28− T-cells in the population of CD8+ T-cells, wherein a percentage of 40% or greater of CD28− T-cells is indicative of an increased risk of cancer.

5. A method of determining an increased risk of cancer in an immunosuppressed patient, the method comprising: determining the percentage of CD8+CD28− T-cells in a sample of CD8+ T-cells from the patient; wherein a percentage of 40% or greater is indicative of an increased risk of cancer.

6. The method according to any preceding claim, wherein the immunosuppression is due to the patient receiving immunosuppressive therapy.

7. The method according to claim 6, wherein the immunosuppressive therapy is following a transplant.

8. The method according to claim 7, wherein the transplant comprises a kidney transplant.

9. The method according to claim 7, wherein the transplant comprises a heart and/or lung transplant.

10. The method according to any of claims 7 to 9, wherein the transplant is not a liver transplant.

11. The method according to any of claims 6 to 9, wherein the immunosuppressive therapy does not comprise a regular and ongoing dosage of steroid.

12. The method according to any preceding claim, comprising the step of selecting patients determined to be at increased risk of cancer for increased surveillance for cancer.

13. The method according to any preceding claim, comprising the step of selecting patients determined to be at increased risk of cancer for modifying the immunosuppressive therapy regime of the patient.

14. The method according to any preceding claim, comprising the step of selecting patients determined to be at increased risk of cancer for providing preventative therapy for cancer.

15. The method according to any of claims 12 to 14, wherein for selected patients determined to be at increased risk of cancer the method further comprises one or more steps comprising: increasing surveillance for cancer; modifying the immunosuppressive therapy regime; or providing preventative therapy for cancer.

16. The method according to any of claims 13 to 15, wherein modifying the immunosuppressive therapy regime comprises one or more of: reduction of dose and/or frequency of immunosuppressive therapy; switching one or more immunosuppressive drugs to an alternative immunosuppressive drug(s); reducing the number of different immunosuppressive drugs administered to the patient.

17. The method according to any of claims 13 to 16, wherein modifying the immunosuppressive therapy regime comprises reduction of dose of immunosuppressive therapy by at least 10% reduction.

18. The method according to any of claims 13 to 17, wherein modifying the immunosuppressive therapy regime comprises reduction of frequency of immunosuppressive therapy.

19. The method according to any of claims 13 to 18, wherein modifying the immunosuppressive therapy regime comprises switching one or more immunosuppressive drugs to an alternative immunosuppressive drug(s).

20. The method according to any of claims 13 to 19, wherein modifying the immunosuppressive therapy regime comprises reducing the number of different immunosuppressive drugs administered to the patient.

21. The method according to any of claims 13 to 20, wherein modification of the immunosuppressive therapy regime is accompanied by an increase in surveillance of transplant rejection.

22. The method according to any preceding claim, wherein the percentage of 40% or greater may be indicative of a risk of cancer after a period of at least 5 years following the onset of immunosuppression in the patient.

23. The method according to any preceding claim, wherein the cancer is skin cancer.

24. The method according to claim 23, wherein the skin cancer is SCC (squamous cell carcinoma).

25. Use of CD57 as a biomarker for determining an increased risk of cancer in a patient receiving immunosuppressive therapy.

26. Use of CD28 as a biomarker for determining an increased risk of cancer in a patient receiving immunosuppressive therapy following a kidney transplant; optionally wherein the cancer is SCC.

27. A kit for determining an increased risk of cancer in a patient receiving immunosuppressive therapy comprising: a CD8 binding agent a CD57 binding agent and/or a CD28 binding agent.

28. A kit for determining an increased risk of cancer in a patient receiving immunosuppressive therapy comprising: a CD4 binding agent a CD57 binding agent and/or a CD28 binding agent.

29. A method, use or kit as substantially described herein, optionally with reference to the accompanying figures.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Embodiments of the invention will now be described in more detail, by way of example only, with reference to the accompanying drawings.

[0096] FIG. 1: Cumulative incidence of non-melanoma skin cancer in Australia (left) and the UK (right—dashed line represents all NMSC whilst solid line represents SCC).

[0097] FIG. 2: Kaplan-Meier survival graph illustrating time from enrolment to SCC development, stratified by history of previous SCC.

[0098] FIG. 3: FIGS. 3A to 3B, like FIG. 2, show survival graphs illustrating time from enrolment to SCC development, stratified by history of previous SCC. In FIG. 3 the days from enrolment are longer. The data presented further demonstrates that the CD57 phenotype predicts SCC development and recurrence. FIG. 3A shows SCC occurrence in all RTRs by history of previous SCC. FIG. 3B shows SCC occurrence in all RTRs by CD57 phenotype. FIG. 3C shows SCC recurrence in all RTRs by time since previous SCC (for those with SCC during study or preceding year). Black ticks indicate censoring. Hazard ratios are reported as univariate Cox models.

[0099] FIG. 4: A) ROC curve for age at enrolment and percentage of CD57+CD8+ cells. RTR was stratified by greater than or less than 50% CD8 cells expressing CD57. B) 2×2 table demonstrating distribution of RTR when stratified by CD57hi or lo. Chi-squared value for the 2×2 table was not significant (p=0.099).

[0100] FIG. 5: A) Kaplan-Meier survival curve demonstrating time to SCC in the highest (CD57hi and previous SCC) and lowest risk groups (CD57lo and no previous SCC). B) Kaplan-Meier survival curve for time between SCC in those with a previous SCC, stratified by CD57hi and CD57lo.

[0101] FIG. 6: IL-2 production in healthy PBMC (n=3) is impaired in CD57+ cells, irrespective of whether the cell is of a naïve or memory phenotype.

[0102] FIG. 7: Shows the proportion of CD8+ T-cells not expressing CD28 (percentage CD8+CD28−) is equally as predictive as (A) and generally correlates strongly with (B) the proportion of CD8+ T-cells expressing CD57 (percentage CD8+CD57+).

[0103] FIG. 8: Demonstrates that CD57 expression upon CD8+ T cells is stable with time and SCC development. Correlation of percentage of CD57-expressing CD8+ T cells between enrolment and 238 days (A) and 385 days (B) later. Points in grey indicate RTRs who developed SCC between sampling. Broken line indicates line of perfect correlation, while solid line indicates line of best fit for whole cohort. Coefficient of determination (r2), standard deviation (SD) and coefficient of variation (CoV) are provided stratified by those who developed SCC between sampling (grey triangles) and those who remained SCC-free (black circles).

DETAILED DESCRIPTION

CD57 Expression as a Predictor of Cutaneous Malignancy in RTR

Introduction

Methods

Recruitment

[0104] Suitable RTR were approached during routine transplant outpatient follow-up at the Oxford Transplant Centre (or its satellite clinics) and asked to participate in the study. Written consent was provided by all participants. Inclusion criteria are detailed in table 1.

TABLE-US-00001 TABLE 1 Recruitment inclusion and exclusion criteria. Inclusion Criteria Exclusion Criteria Male and female RTR aged greater than 18 Unable/unwilling to provide informed years old. consent to participate A stable, functioning renal transplant Previous invasive malignancy during the (defined as serum creatinine increased last five years (except cutaneous basal cell <30% above previous value in preceding carcinoma*) 12 weeks). Have provided informed consent to Evidence of systemic infection at time of participate recruitment (see below) Transplant recipient of any organ other than kidney previously Previous evidence of human immunodeficiency virus (HIV) infection First cSCC developed within 1 year of first transplant *or cutaneous squamous cell carcinoma in group 1.

[0105] RTR with and without a history of prior cutaneous SCC were matched by age, sex and total duration of immunosuppression. A questionnaire was completed regarding sun exposure and risk factors for malignancy development. Renal function was assessed by the serum creatinine performed most recently to time of sampling (in the vast majority this was on the same day) and by calculation of the eGFR (using the four-variable MDRD equation). For the purpose of determining cSCC-free survival, time until occurrence was taken as from the date of recruitment until the date of excision of a histologically-confirmed cSCC. For the purposes of counting discrete cSCC, an event was not counted if it was considered a recurrence; either by description in the medical notes, or report of a scar within the histology (with previous excision at that site within one year). Where there was diagnostic uncertainty (e.g. keratoacanthoma versus cSCC), the reporting histologist's opinion was taken as final. If no opinion was given as to the more likely diagnosis, the event was not included.

Clinical Risk Scores

[0106] Three risk scores for use in transplant patients have been published previously, and all three were calculated for all participants. These were: [0107] Harwood risk score[8] as defined by:

TABLE-US-00002 Risk score Phenotype 1 Fitzgerald skin type 5-6; 2 Fitzgerald skin type 1-4, <35 years old at first transplant; 3 Fitzgerald skin type 1-4, 35-44 years old at first transplant, <5 lifetime sunburns; 4 As per risk score 3 but >5 lifetime sunburns or 45-54 years old at first transplant; 5 Skin type 1-4 and greater than 55 years old at first transplant. [0108] Urwin risk score[9] as calculated by the sum of 2 points if age >50 at first transplant; 2 points if average daily lifetime exposure to sunlight >1 hour; 2 points if >30 years in a tropical climate; 3 points if the participant had an SCC prior to transplant; 2 points if the participant had another NMSC prior to transplant; 1 point for any history of childhood sunburn (taken to be <16 years old); 1 point for Fitzgerald skin type 1. [0109] Harden risk score[7] as determined by the risk score (M×1.26)+(A×0.1)+(G×1.87) where M is gender (where male scores 1, female scores 0), A is age at transplantation (in years) and G is eye colour (where green eyes score 1, all other colours score 0).

Flow Cytometric Analysis

[0110] Blood for the study was taken at the same time as routine clinical venepuncture, then immediately stored on ice. All samples were processed within four hours of venepuncture and was anticoagulated with EDTA in vacutainers (BD Biosciences, Oxford, UK).

[0111] Peripheral blood mononuclear cells (PBMC) were separated using density-gradient centrifugation with Lymphocyte Separation Medium (GE Healthcare, Amersham, UK). Fresh PBMC were then incubated with a cocktail of monoclonal antibodies including CD3, CD57, CD28 (eBiosciences, Hatfield, UK), CD8 (BD Biosciences, UK) and CD4 (Beckman Coulter, UK) for 45 minutes at 4° C. The stained cells were then analysed using the Navios flow cytometry system (Beckman Coulter).

Data Analysis

[0112] Flow cytometry analysis was performed using Kaluza version 1.2 (Beckman Coulter, UK). Statistical analysis was performed using SPSS 20 (IBM Corp, NY) or Graphpad Prism for Windows version 5.03 (Graphpad, San Diego, Calif.). Results are shown as median (interquartile range) unless specified otherwise, except hazard and odds ratios which are reported as hazard/odds ratio (95% confidence interval). Comparison between groups was performed using the non-parametric two-tailed Mann-Whitney (two groups) or Kruskal-Wallis (multiple groups) test, with continuous variables. For categorical variables the chi-squared or Fisher's exact test were used as appropriate. Where the Kruskal-Wallis test was performed, a subsequent post-hoc Dunn test was applied. Correlations were tested using Pearson's test. Multivariate Cox regression was performed using a backward stepwise method. A p-value of less than 0.05 was considered significant. The statistics presented in this document represent those performed by the author with review by Cristian Ciria and Sharon Love of the Centre for Statistics in Medicine, University of Oxford.

[0113] The study received a favourable ethical opinion from the NHS Research Ethics Committee (reference: 12/WS/0288).

Results (NB Data Collection is Ongoing—Results Accurate as of 13 Oct. 2014)

Recruitment and Demographics

[0114] 116 RTR have been recruited to date and the demographics of the group are summarised in Table 2. Of note, those with a history of SCC are older and with a trend towards a lower BMI than those without a history of SCC. Amongst risk factors for SCC, those with a history of SCC were significantly more likely to have a first-degree family history of malignancy, and had a significantly increased Urwin risk score compared to those without a history of SCC (table 2). Retrospectively, RTR with a history of SCC had a trend towards an increased proportion of CD8+ cells expressing CD57; there was no difference in absolute number.

TABLE-US-00003 TABLE 2 study participant demographics & clinical phenotype. A smoker was defined as a participant with >1 pack-year history of smoking. Family history of malignancy was defined as malignancy in a sibling or parent. Chronic UV exposure was defined as by employment in an outdoor occupation for greater than five years; continuous residence in a sunny climate for greater than six months; or more than 11 holidays to a sunny climate where the participant sunbathed[8]. SCC No SCC p Number  59  57 % male 71% 67% 0.60 % CMV seropositive at enrolment 68% 63% 0.55 Median (IQR) age (yr) at enrolment  66  61 0.02 (58-74) (55-67) Median (IQR) age (yr) at 1st Tx  43  40 0.26 (31-52) (32-47) Median (IQR) body mass index (kg/m.sup.2) at  25.2  26.1 0.08 enrolment (21.7-28.2) (23.3-29.7) Median (IQR) i'suppression time (mo) at 283 249 0.16 enrolment (208-353) (203-314) Median (IQR) number of transplants  1  1 0.59 (1-1) (1-1) Median (IQR) serum creatinine at 117 128 0.24 enrolment (92-165) (108-159) Median (IQR) eGFR at enrolment  51  44 0.16 (36-64) (34-57) Immunosuppression at enrolment % on calcineurin inhibitor 83% 81% 0.74 % on azathiaprine 78% 65% 0.12 % on mycophenolate  7% 16% 0.12 % on sirolimus  7%  4% 0.68 % on steroids 41% 39% 0.82 Clinical phenotype % smoker (past or current) 34% 39% 0.60 % family history of malignancy 53% 35% 0.04 % personal history of non-NMSC  9% 19% 0.16 malignancy % reporting chronic UV exposure 64% 56% 0.36 Median (IQR) Harwood clinical risk score  4  3 0.09 (2-4) (2-4) Median (IQR) Urwin clinical risk score  2  1 0.01 (1-3) (0-3) Median (IQR) Harden clinical risk score  5.3  4.9 0.38 (4.2-6.2) (4.3-5.7) Immune phenotype Number of CD8+ CD57+ cells (per ul 123 145 0.45 blood) (34-233) (76-304) Median (IQR) % CD8+ expressing CD57  67  49 0.13 (41-76)% (35-70)%

Prediction of SCC Development Using Clinical Markers

[0115] Participants in the study were followed up for a median (IQR) period of 371 (285-472) days, representing a total of 34890 days ‘at risk’ of SCC. During the follow up period 22 RTR developed a total of 36 SCC. RTR with a history of SCC were at nearly four-fold risk of further SCC, as demonstrated in FIG. 2.

[0116] A univariate analysis was performed to assess clinical factors predictive of further SCC development in this cohort and it was found that age at recruitment and at first transplantation, as well as history of previous SCC, were predictive of further SCC development, as detailed in Table 3. Gender, duration of immunosuppression, CMV serostatus, renal function and immunosuppression type were not predictive of SCC development (data not shown). Only the Urwin risk score was found to be predictive of further SCC development, though both the Harwood and Harden risk scores trended towards significance.

[0117] Using multivariate analysis to account for potential confounding between factors, it was observed that, correcting for age, all three risk scores lost predictive value. Furthermore, age at first transplant also lost predictive value, leaving age at enrolment and previous SCC as the only independently predictive clinical markers for SCC development.

TABLE-US-00004 TABLE 3 Regression analyses for SCC development during the study using clinical markers. Multivariate analysis was performed using previous SCC, age at transplantation and age at enrolment as covariates, whilst each individual risk score was assessed using age as a covariate. For continuous variables, such as age, the hazard ratio is per unit (e.g. year) increase. Univariate analysis Hazard Ratio (HR) Multivariate analysis* Variable (95% CI) p HR (95% CI) p Increasing age at 1.07 (1.02-1.11) 0.002 1.05 (1.01-1.09) 0.016 enrolment Increasing age at 1.sup.st 1.03 (1.00-1.07) 0.050 0.99 (0.94-1.04) 0.648 transplant Previous SCC 5.48 (1.85-16.2) 0.002 4.39 (1.45-13.3) 0.009 Chronic UV exposure 2.75 (1.01-7.47) 0.047 2.65 (0.96-7.36) 0.06 Clinical risk scores ↑ Harden risk score 1.31 (0.99-1.7)  0.06 1.01 (0.71-1.46) 0.94 ↑ Urwin risk score 1.50 (1.12-2.02) 0.007 1.31 (0.95-1.80) 0.10 ↑ Harwood risk score 1.43 (0.97-2.12) 0.07 0.84 (0.47-1.50) 0.55

Prediction of SCC Using Percentage of CD8+ Cells Expressing CD57

[0118] Having established the performance of clinical risk markers in this population, their performance was compared with the percentage of CD57+ cells as predictors of further SCC. A receiver operating characteristic (ROC) curve (FIG. 4) was produced. This gave a similar predictive performance of CD57 and age, as assessed by area under the curve. It was observed that CD57 percentage was bimodal (largely due to the influence of CMV serostatus) and which allow the RTR to be stratified into those with a majority (>50%) of CD8+ cells expressing CD57 (CD57.sup.hi) and those with a minority)(CD57.sup.lo. RTR with a history of SCC tended to be in the CD57hi group, though this did not reach significance (FIG. 3). ROC curves were also construced with regards to BCC development (n=13) and other malignancy (n=3), but this was not predictive.

[0119] Using this stratification, it was found on univariate analysis that RTR who were CD57hi were over four times more likely to develop SCC during follow-up than CD57lo RTR (Table 4). When corrected for age, only a history of previous SCC and being CD57hi remained predictive of the development of SCC. CD57hi status remained predictive of the development of SCC, independent of previous SCC and increasing age. Given the impact of CMV seropositivity on the proportion of CD57+ cells (CMV seropositive RTR were eighteen times more likely to be CD57hi), further investigation into this relationship was undertaken but it was found on multivariate analysis that CD57 remained independently predictive of SCC development (HR 4.4 (1.49-13.1), p=0.007) when adjusted for CMV serostatus.

TABLE-US-00005 TABLE 4 Regression analysis using CD57hi and CD57lo to stratify. All three variables were used as covariates for multivariate analysis. Univariate Multivariate Variable HR (95% CI) p HR (95% CI) p ↑ age at recruitment 1.07 0.002 1.03 0.14 (1.02-1.11) (0.99-1.08) Previous SCC 5.48 0.002 4.9 0.005 (1.85-16.2) (1.63-14.5) >50% CD8 expressing CD57 4.45 0.002 3.9 0.015 (CD57hi) (1.50-13.2) (1.30-11.5)

[0120] RTR who were highest risk (previous SCC and CD57hi) were nearly seven-fold more likely to develop an SCC during study follow-up compared to those who were lowest risk (FIG. 5A). Furthermore, when time since previous SCC was examined in those who were CD57hi compared to those who were CD57lo, it was observed that those who were CD57hi trended towards earlier SCC recurrence (FIG. 5B), and were three-fold more likely to develop an SCC during study follow-up (HR 3.01 (1.17-7.73), p=0.02, data not shown).

[0121] The data presented in FIG. 8 demonstrates that CD57 is stable for over a 1-year period and with SCC development, further supporting its usefulness as a marker.

Immunological Features of CD8+CD57+ Cells

[0122] CD8+CD57+ cells were further analysed ex vivo. These cells were predominantly derived from effector memory populations, and were markedly impaired in the production of IL-2 in response to polyclonal stimulation in healthy individuals (FIG. 6). Furthermore, an increased proportion of CD8+CD57+ cells was associated with a reduction in the number of naïve and central memory CD8 T-cells. The percentage of CD8+CD57+ cells correlates with the percentage of CD4+CD57+ cells, which are also impaired in the ability to produce IL-2.

CONCLUSIONS & CONTEXT

[0123] This study is the first to directly compare the performance of clinical risk markers against immunological risk markers in the prediction of SCC in long-term RTR, who can generally be considered to be at globally ‘high risk’ due to cumulative time of immunosuppression and increased age. The proportion of CD57-expressing CD8+ T-cells represents the first immunological marker that has been demonstrated to have superior predictive power beyond traditionally utilised clinical indicators, in this setting.

[0124] The identification of a marker that allows stratification of RTR with regards to risk of SCC has a number of implications. Firstly, this may allow for more intensive screening in those who are high risk, whilst those at lower risk may require less intensive follow-up. This potentially could free dermatological resources to monitor and intervene with those at highest risk of SCC. Secondly, CD8+CD57+ cells are thought to represent an ‘exhausted’ phenotype; data supported by the findings presented here. It may well be that the immune system's impaired response to the development of malignancy is indicative of an impaired response to the graft itself, which may underlie the ability to frequently (though not invariably) reduce immunosuppression in these patients without the development of graft dysfunction. Stratification by peripheral blood CD57 proportion may act as a biomarker to identify those RTR that may be able to reduce their immunosuppression prior to the development of malignancy. An additional benefit is a financial one: standard double therapy immunosuppression (i.e. ciclosporin and azathioprine) costs in the region of £1800 per annum per patient (A. Devaney, personal communication). A reduction in dosage of these therapies in a subset of RTR may have ongoing cost-saving implications for the healthcare provider.

[0125] CD57 has been shown to be a poor prognostic marker in a number of malignancies, including gastric, melanoma and renal cell. However, these studies looked only at participants who had malignancy at the time of sampling. Thus these studies did not identify CD57 in the context of predicting de novo malignancy development, but rather progression of established malignancy.

[0126] CD57 has been previously investigated in the setting of both transplantation and malignancy. Boleslawski (2011) looked at a cohort of liver transplant recipients and found that a decreased proportion of CD8+CD28+(and by inversion an increased proportion of CD8+CD28−) cells predicted the development of malignancy in the first ten years post-transplant[32]. CD8+CD57+ cells are often (though not always) CD28-negative (a finding confirmed by the inventors with a strong correlation between these two populations). Liver transplantation generally requires much lower doses of immunosuppression as tolerance is a more common phenomenon (up to 30% of liver transplant recipients are able to cease immunosuppression) in comparison to kidney transplantation (where tolerance is thought to be very rare) and thus the types of malignancy encountered post-transplant are likely to differ, with a relative overrepresentation of liver cancer (which isn't seen in renal transplantation). Only a third of the malignancies encountered during follow-up in the Boleslawski study were skin malignancies, and these were experienced by only 6% of the study cohort, highlighting that skin malignancy is generally encountered much later in the post-transplant course. Courivaud analysed malignancy in the first decade following renal transplantation and found that CMV seropositive RTR were more likely to develop malignancy than those who were CMV seronegative. In addition, this study found that CMV seropositivity was associated with an increased percentage of CD8+CD57+IL-2− cells; however, they did not use CD57 itself as a predictive marker for malignancy. The inventors also have observed that CMV seropositivity is associated with an eighteen-fold increased likelihood of being CD57hi, but CMV seropositivity itself was not predictive of SCC development, whilst the proportion of CD57+ cells remained independently predictive. Given the data above, it is likely that CD57 is predictive of the development of other malignancies in long-term RTR, but the paucity of these events in this study means we are unable to formally assess this.

[0127] The poor performance of clinical risk scores may be due to a number of reasons. The Harden clinical risk score was developed in a cohort of RTR who were in the first 10 years post-transplant, where skin cancer is relatively underrepresented. Secondly, the populations in this study may differ from those where the measures were developed. The Urwin risk score was developed utilising an Australian population, and so gives relative prominence to features such as duration of time in a tropical climate and pre-transplant NMSC, both of which are unusual in a British cohort.

REFERENCES

[0128] 1. Survival Rates Following Transplantation. [cited 2014 16 Apr.]; Available from: www.organdonation.nhs.uk/ukt/statistics/transplant_activity_report/current_activity_reports/ukt/survival_rates_following_transplantation.pdf. [0129] 2. Matas, A. J., et al., OPTN/SRTR 2011 Annual Data Report: kidney. Am J Transplant, 2013. 13 Suppl 1: p. 11-46. [0130] 3. Pilmore, H., et al., Reduction in cardiovascular death after kidney transplantation. Transplantation, 2010. 89(7): p. 851-7. [0131] 4. Euvrard, S., et al., Subsequent skin cancers in kidney and heart transplant recipients after the first squamous cell carcinoma. Transplantation, 2006. 81(8): p. 1093-100. [0132] 5. Baker, R., A. Jardine, and P. Andrews. Post-operative Care of the Kidney Transplant Recipient. 2011 5 Feb. 2011 [cited 2014 17 Oct.]; Available from: http://www.renal.org/guidelines/modules/post-operative-care-of-the-kidney-transplant-recipient#sthash.OIcp7tZp.dpbs. [0133] 6. US Renal Data System. USRDS 2012 Annual Data Report: Atlas of End-Stage Renal Disease in the United States. 2012, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Md. [0134] 7. Harden, P. N., et al., Annual incidence and predicted risk of nonmelanoma skin cancer in renal transplant recipients. Transplant Proc, 2001. 33(1-2): p. 1302-4. [0135] 8. Harwood, C. A., et al., A Surveillance Model for Skin Cancer in Organ Transplant Recipients: A 22-Year Prospective Study in an Ethnically Diverse Population. Am J Transplant, 2012. [0136] 9. Urwin, H. R., et al., Predicting risk of nonmelanoma skin cancer and premalignant skin lesions in renal transplant recipients. Transplantation, 2009. 87(11): p. 1667-71. [0137] 10. Dantal, J., et al., Effect of long-term immunosuppression in kidney-graft recipients on cancer incidence: randomised comparison of two cyclosporin regimens. Lancet, 1998. 351(9103): p. 623-8. [0138] 11. Carroll, R. P., et al., Immune phenotype predicts risk for posttransplantation squamous cell carcinoma. J Am Soc Nephrol, 2010. 21(4): p. 713-22. [0139] 12. Hope, C. M., et al., The immune phenotype may relate to cancer development in kidney transplant recipients. Kidney Int, 2014. [0140] 13. Sherston, S. N., et al., Demethylation of the TSDR Is a Marker of Squamous Cell Carcinoma in Transplant Recipients. Am J Transplant, 2014. [0141] 14. Kosmidis, M., et al., Immunosuppression affects CD4+ mRNA expression and induces Th2 dominance in the microenvironment of cutaneous squamous cell carcinoma in organ transplant recipients. J Immunother, 2010. 33(5): p. 538-46. [0142] 15. Muhleisen, B., et al., Progression of cutaneous squamous cell carcinoma in immunosuppressed patients is associated with reduced CD123+ and FOXP3+ cells in the perineoplastic inflammatory infiltrate. Histopathology, 2009. 55(1): p. 67-76. [0143] 16. Zhang, S., et al., Increased Tc22 and Treg/CD8 ratio contribute to aggressive growth of transplant associated squamous cell carcinoma. PLoS One, 2013. 8(5): p. e62154. [0144] 17. Strioga, M., V. Pasukoniene, and D. Characiejus, CD8+ CD28− and CD8+ CD57+ T cells and their role in health and disease. Immunology, 2011. 134(1): p. 17-32. [0145] 18. Cebo, C., et al., Function and molecular modeling of the interaction between human interleukin 6 and its HNK-1 oligosaccharide ligands. J Biol Chem, 2002. 277(14): p. 12246-52. [0146] 19. Nielsen, C. M., et al., Functional Significance of CD57 Expression on Human NK Cells and Relevance to Disease. Front Immunol, 2013. 4: p. 422. [0147] 20. Brenchley, J. M., et al., Expression of CD57 defines replicative senescence and antigen-induced apoptotic death of CD8+ T cells. Blood, 2003. 101(7): p. 2711-20. [0148] 21. Chong, L. K., et al., Proliferation and interleukin 5 production by CD8hi CD57+ T cells. Eur J Immunol, 2008. 38(4): p. 995-1000. [0149] 22. Akagi, J. and H. Baba, Prognostic value of CD57(+) T lymphocytes in the peripheral blood of patients with advanced gastric cancer. Int J Clin Oncol, 2008. 13(6): p. 528-35. [0150] 23. Characiejus, D., et al., Peripheral blood CD8highCD57+ lymphocyte levels may predict outcome in melanoma patients treated with adjuvant interferon-alpha. Anticancer Res, 2008. 28(2B): p. 1139-42. [0151] 24. Characiejus, D., et al., Predictive value of CD8highCD57+ lymphocyte subset in interferon therapy of patients with renal cell carcinoma. Anticancer Res, 2002. 22(6B): p. 3679-83. [0152] 25. Filaci, G., et al., CD8+ CD28− T regulatory lymphocytes inhibiting T cell proliferative and cytotoxic functions infiltrate human cancers. J Immunol, 2007. 179(7): p. 4323-34. [0153] 26. Focosi, D., et al., CD57+ T lymphocytes and functional immune deficiency. J Leukoc Biol, 2010. 87(1): p. 107-16. [0154] 27. Hadrup, S. R., et al., Longitudinal studies of clonally expanded CD8 T cells reveal a repertoire shrinkage predicting mortality and an increased number of dysfunctional cytomegalovirus-specific T cells in the very elderly. J Immunol, 2006. 176(4): p. 2645-53. [0155] 28. Hebib, C., et al., Pattern of cytokine expression in circulation CD57+ T cells from long-term renal allograft recipients. Transpl Immunol, 1998. 6(1): p. 39-47. [0156] 29. Le Priol, Y., et al., High cytotoxic and specific migratory potencies of senescent CD8+ CD57+ cells in HIV-infected and uninfected individuals. J Immunol, 2006. 177(8): p. 5145-54. [0157] 30. Sadat-Sowti, B., et al., An inhibitor of cytotoxic functions produced by CD8+CD57+ T lymphocytes from patients suffering from AIDS and immunosuppressed bone marrow recipients. Eur J Immunol, 1994. 24(11): p. 2882-8. [0158] 31. Wang, E. C., et al., CD8high+ (CD57+) T cells in patients with rheumatoid arthritis. Arthritis Rheum, 1997. 40(2): p. 237-48. [0159] 32. Boleslawski, E., et al., CD28 expression by peripheral blood lymphocytes as a potential predictor of the development of de novo malignancies in long-term survivors after liver transplantation. Liver Transpl, 2011. 17(3): p. 299-305.