Original Clinical Science
Human Leukocyte Antigen Type and Posttransplant Lymphoproliferative Disorder Mark E. Lustberg,1 Ronald P. Pelletier,2 Pierluigi Porcu,3,4 Stanley I. Martin,1 Carl D. Quinion,4 Susan M. Geyer,3 Michael A. Caligiuri,3,4 and Robert A. Baiocchi3,4
Background. Posttransplant lymphoproliferative disorder (PTLD) is an infrequent but serious complication of solid organ trans-
plantation. Early detection and initiation of therapy may improve outcomes. The purpose of this study was to identify human leukocyte antigen (HLA) type as risk and prognostic factors for PTLD. Methods. A review was undertaken to identify PTLD cases treated at our institution over the past 25 years. Logistic regression and Cox Proportional Hazards were used to model risk factors for PTLD and clinical outcomes in patients with PTLD. Results. One hundred six cases of PTLD were identified with 1392 solidorgan transplant recipient controls. Epstein-Barr virus (EBV) seronegative status pretransplant (odds ratio [OR] = 7.61, 95% confidence interval [95% CI] = 3.83-15.1) and receipt of a nonkidney transplant were associated with an increased risk of PTLD. Being African American and receipt of a living-related kidney transplant were associated with a decreased risk of PTLD. The HLA-B40 group was a risk factor for PTLD in EBV-seronegative individuals (OR = 8.38, 95% CI = 2.18-32.3), whereas HLA-B8 was a risk factor for PTLD in EBV-seropositive individuals (OR = 3.29, 95% CI = 1.52-7.09). Specific HLA types were not associated with graft failure or mortality after PTLD diagnosis. In PTLD patients, central nervous system (CNS) involvement, bone marrow involvement, T-cell PTLD, and age were associated with increased mortality. Conclusion. Human leukocyte antigen-B40 group and HLA-B8 were identified as novel susceptibility factors for PTLD in EBV-seropositive and EBV-seronegative individuals, respectively. Multicentered, large prospective studies of PTLD with correlative immunologic work are needed to test the significance of these observed associations. (Transplantation 2015;99: 1220–1225)
S
olid organ transplant recipients are profoundly immune suppressed and are at risk for developing post-transplant lymphoproliferative disorder (PTLD). Affecting 1% to 4% of transplant patients,1-7 PTLD is associated with substantial morbidity and mortality.8,9 Approximately 80% of PTLD cases are related to infection with the Epstein-Barr virus (EBV),10 a ubiquitous, oncogenic herpes virus.10-17 In immune competent individuals, EBV infected B cells are eliminated by an efficient cellular immune response, preventing the development of EBV-associated lymphoproliferative disorders. However, in patients receiving immune suppressive
Received 28 October 2013. Revision requested 18 November 2013.
medications in the posttransplant setting PTLD can develop. Unlike other lymphoid neoplasms that have shown improved outcomes over the last decade, PTLD remains a disease with a poor prognosis and inadequate treatment options.5,8,18-22 Thus, early disease detection and initiation of therapy may be critical for improving outcomes. Strategies are needed to risk stratify transplant patients for the development of PTLD and to identify patients who will likely benefit most from more aggressive initial therapy (i.e., chemotherapy). Although clinical risk factors and prognostic risk factors for PTLD have been described,2,3,9,23 additional factors need to be identified to more precisely predict disease and outcomes in patients with PTLD. Previous reports suggest that immunogenetic variables may be risk factors for PTLD.1,24-39
Accepted 2 September 2014. 1 Division of Infectious Diseases, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH. 2 Division of Transplantation Surgery, Department of Surgery, The Ohio State University Wexner Medical Center, Columbus, OH. 3 Division of Hematology, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH. 4 The Ohio State University Comprehensive Cancer Center, Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, Columbus, OH.
research, and writing the article. P.P. participated in research design and in writing the article. S.I.M. participated in research design and in writing the article. C.D.Q. participated in performance of the research and writing the article. S.M.G. provided statistical guidance on the article revision. M.A.C. participated in research design and in writing the article. R.A.B. participated in research design, performance of the research, data analysis, and writing the article. Correspondence: Robert A. Baiocchi, MD, PhD, B304 Starling Loving Hall, 320 West 10th Avenue, Columbus, OH 43210. ([email protected]).
Supporting information: NIH T32CA090223 (M.E.L.). American Cancer Society Institutional Research Seed Grant (M.E.L.). Pelotonia Research Foundation (M.E.L.). Friends of Jason Gould Foundation (R.A.B.).
Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com).
The authors declare no conflicts of interest.
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
M.E.L. participated in research design, performance of the research, data analysis, and writing the article. R.P.P. participated in research design, performance of the
ISSN: 0041-1337/15/9906-1220
1220
www.transplantjournal.com
DOI: 10.1097/TP.0000000000000487
Transplantation
■
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
June 2015
■
Volume 99
■
Number 6
Lustberg et al
© 2014 Wolters Kluwer
The purpose of the current retrospective study is to characterize human leukocyte antigen (HLA) types as risk factors for PTLD and prognostic factors in patients with PTLD. RESULTS Factors Associated With PTLD
The cohort consisted of 1,498 individuals, with 106 cases of PTLD. The majority of PTLD cases were EBV positive, with 25 of 106 cases (23.6%) staining negative for EBV. As indicated in Table 1, male sex was significantly associated with EBV-positive PTLD (odds ratio [OR] = 2.09, 95% confidence interval [95% CI] = 1.22-3.60, P = 0.007), but not EBV-negative PTLD. The EBV-seronegative status pretransplant and receipt of a nonkidney transplant (liver, heart, pancreas, or lung) were major risk factors for PTLD (OR = 7.61, 95% CI = 3.83-15.1, P < 0.001 and OR = 2.99 (95% CI = 1.92-4.66, P < 0.001, respectively). Being African American and receipt of a living-related kidney transplant were significantly associated with a decreased risk of PTLD (OR = 0.29, 95% CI = 0.13-0.63, P = 0.002 and OR = 0.30, 95% CI = 0.16-0.53, P < 0.001, respectively). We evaluated the effects of HLA-type (HLA-A, HLA-B, and HLA-DR) as risk factors for PTLD in univariable and
1221
multivariable analyses (Fig. 1, Table 2, and Supplemental Tables, SDC, http://links.lww.com/TP/B80). We found that HLA-B40 group was an independent risk factor for PTLD, OR = 2.62, 95% CI = 1.59–4.31, uncorrected P = 0.00016. This P value is highly statistically significant, even after correction for multiple comparisons, Bonferroni adjusted P value = 0.006. Adjusting for multiple comparisons none of the other HLA types were significantly associated with PTLD in the cohort as a whole. Given the strong association of baseline EBV-seronegative status with PTLD, we undertook exploratory analyses by baseline EBV serostatus. In subgroup analyses in those with EBV-negative serostatus at baseline, HLA-B40 group was associated with an OR = 8.38 (95% CI = 2.18-32.3, P = 0.002) for PTLD. In those with EBV-seropositive status at baseline, HLA-B8 was associated with PTLD, OR = 3.29 (95% CI = 1.52-7.09, P = 0.0024). Individuals with PTLD in this study had significantly longer follow-up time than those without PTLD (Table 1). To address this, Cox proportional hazards models were used to model time to the development of PTLD. As can be seen in Table 2, effect estimates were generally similar between the multivariable logistic model and the multivariable Cox proportional hazards model. To determine if year of transplant
TABLE 1.
Characteristics of controls, EBV-positive PTLD cases, and EBV-negative PTLD cases Descriptive characteristics
Controls
Age (mean ± SD), yr Male sex African American EBV-seronegative preransplantb Type of transplantc LRD kidney LUD kidney Deceased donor kidney Kidney + pancreasd Other—liver, heart, lung, pancreas aloned Reason for transplante Diabetes Hypertension Glomerulonephritis Study follow up time (mean ± SD), yr Detailed characteristics of PTLD cases Years from transplant to diagnosis (median, IQR) B-cell PTLD T-cell PTLD Documented CNS involvement Documented marrow involvement Lymphoma on pathology/monomorphic PTLD Graft failure after PTLD Removal of graft (nephrectomy or pancreatectomy) Median survival years after diagnosis
Pa
EBV-positive PTLD
EBV-negative PTLD
45.8 ± 14.9 60.1% (836/1392) 19.8% (275/1392) 6.38% (73/1144)
45.6 ± 14.6 76.5% (62/81) 8.64% (7/81) 40.0% (14/35)
47.5 ± 13.2 60.0% (15/25) 0% (0/25) 0% (0/6)
0.83 0.013 0.002
Human Leukocyte Antigen Type and Posttransplant Lymphoproliferative Disorder Mark E. Lustberg,1 Ronald P. Pelletier,2 Pierluigi Porcu,3,4 Stanley I. Martin,1 Carl D. Quinion,4 Susan M. Geyer,3 Michael A. Caligiuri,3,4 and Robert A. Baiocchi3,4
Background. Posttransplant lymphoproliferative disorder (PTLD) is an infrequent but serious complication of solid organ trans-
plantation. Early detection and initiation of therapy may improve outcomes. The purpose of this study was to identify human leukocyte antigen (HLA) type as risk and prognostic factors for PTLD. Methods. A review was undertaken to identify PTLD cases treated at our institution over the past 25 years. Logistic regression and Cox Proportional Hazards were used to model risk factors for PTLD and clinical outcomes in patients with PTLD. Results. One hundred six cases of PTLD were identified with 1392 solidorgan transplant recipient controls. Epstein-Barr virus (EBV) seronegative status pretransplant (odds ratio [OR] = 7.61, 95% confidence interval [95% CI] = 3.83-15.1) and receipt of a nonkidney transplant were associated with an increased risk of PTLD. Being African American and receipt of a living-related kidney transplant were associated with a decreased risk of PTLD. The HLA-B40 group was a risk factor for PTLD in EBV-seronegative individuals (OR = 8.38, 95% CI = 2.18-32.3), whereas HLA-B8 was a risk factor for PTLD in EBV-seropositive individuals (OR = 3.29, 95% CI = 1.52-7.09). Specific HLA types were not associated with graft failure or mortality after PTLD diagnosis. In PTLD patients, central nervous system (CNS) involvement, bone marrow involvement, T-cell PTLD, and age were associated with increased mortality. Conclusion. Human leukocyte antigen-B40 group and HLA-B8 were identified as novel susceptibility factors for PTLD in EBV-seropositive and EBV-seronegative individuals, respectively. Multicentered, large prospective studies of PTLD with correlative immunologic work are needed to test the significance of these observed associations. (Transplantation 2015;99: 1220–1225)
S
olid organ transplant recipients are profoundly immune suppressed and are at risk for developing post-transplant lymphoproliferative disorder (PTLD). Affecting 1% to 4% of transplant patients,1-7 PTLD is associated with substantial morbidity and mortality.8,9 Approximately 80% of PTLD cases are related to infection with the Epstein-Barr virus (EBV),10 a ubiquitous, oncogenic herpes virus.10-17 In immune competent individuals, EBV infected B cells are eliminated by an efficient cellular immune response, preventing the development of EBV-associated lymphoproliferative disorders. However, in patients receiving immune suppressive
Received 28 October 2013. Revision requested 18 November 2013.
medications in the posttransplant setting PTLD can develop. Unlike other lymphoid neoplasms that have shown improved outcomes over the last decade, PTLD remains a disease with a poor prognosis and inadequate treatment options.5,8,18-22 Thus, early disease detection and initiation of therapy may be critical for improving outcomes. Strategies are needed to risk stratify transplant patients for the development of PTLD and to identify patients who will likely benefit most from more aggressive initial therapy (i.e., chemotherapy). Although clinical risk factors and prognostic risk factors for PTLD have been described,2,3,9,23 additional factors need to be identified to more precisely predict disease and outcomes in patients with PTLD. Previous reports suggest that immunogenetic variables may be risk factors for PTLD.1,24-39
Accepted 2 September 2014. 1 Division of Infectious Diseases, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH. 2 Division of Transplantation Surgery, Department of Surgery, The Ohio State University Wexner Medical Center, Columbus, OH. 3 Division of Hematology, Department of Internal Medicine, The Ohio State University Wexner Medical Center, Columbus, OH. 4 The Ohio State University Comprehensive Cancer Center, Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, Columbus, OH.
research, and writing the article. P.P. participated in research design and in writing the article. S.I.M. participated in research design and in writing the article. C.D.Q. participated in performance of the research and writing the article. S.M.G. provided statistical guidance on the article revision. M.A.C. participated in research design and in writing the article. R.A.B. participated in research design, performance of the research, data analysis, and writing the article. Correspondence: Robert A. Baiocchi, MD, PhD, B304 Starling Loving Hall, 320 West 10th Avenue, Columbus, OH 43210. ([email protected]).
Supporting information: NIH T32CA090223 (M.E.L.). American Cancer Society Institutional Research Seed Grant (M.E.L.). Pelotonia Research Foundation (M.E.L.). Friends of Jason Gould Foundation (R.A.B.).
Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com).
The authors declare no conflicts of interest.
Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
M.E.L. participated in research design, performance of the research, data analysis, and writing the article. R.P.P. participated in research design, performance of the
ISSN: 0041-1337/15/9906-1220
1220
www.transplantjournal.com
DOI: 10.1097/TP.0000000000000487
Transplantation
■
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
June 2015
■
Volume 99
■
Number 6
Lustberg et al
© 2014 Wolters Kluwer
The purpose of the current retrospective study is to characterize human leukocyte antigen (HLA) types as risk factors for PTLD and prognostic factors in patients with PTLD. RESULTS Factors Associated With PTLD
The cohort consisted of 1,498 individuals, with 106 cases of PTLD. The majority of PTLD cases were EBV positive, with 25 of 106 cases (23.6%) staining negative for EBV. As indicated in Table 1, male sex was significantly associated with EBV-positive PTLD (odds ratio [OR] = 2.09, 95% confidence interval [95% CI] = 1.22-3.60, P = 0.007), but not EBV-negative PTLD. The EBV-seronegative status pretransplant and receipt of a nonkidney transplant (liver, heart, pancreas, or lung) were major risk factors for PTLD (OR = 7.61, 95% CI = 3.83-15.1, P < 0.001 and OR = 2.99 (95% CI = 1.92-4.66, P < 0.001, respectively). Being African American and receipt of a living-related kidney transplant were significantly associated with a decreased risk of PTLD (OR = 0.29, 95% CI = 0.13-0.63, P = 0.002 and OR = 0.30, 95% CI = 0.16-0.53, P < 0.001, respectively). We evaluated the effects of HLA-type (HLA-A, HLA-B, and HLA-DR) as risk factors for PTLD in univariable and
1221
multivariable analyses (Fig. 1, Table 2, and Supplemental Tables, SDC, http://links.lww.com/TP/B80). We found that HLA-B40 group was an independent risk factor for PTLD, OR = 2.62, 95% CI = 1.59–4.31, uncorrected P = 0.00016. This P value is highly statistically significant, even after correction for multiple comparisons, Bonferroni adjusted P value = 0.006. Adjusting for multiple comparisons none of the other HLA types were significantly associated with PTLD in the cohort as a whole. Given the strong association of baseline EBV-seronegative status with PTLD, we undertook exploratory analyses by baseline EBV serostatus. In subgroup analyses in those with EBV-negative serostatus at baseline, HLA-B40 group was associated with an OR = 8.38 (95% CI = 2.18-32.3, P = 0.002) for PTLD. In those with EBV-seropositive status at baseline, HLA-B8 was associated with PTLD, OR = 3.29 (95% CI = 1.52-7.09, P = 0.0024). Individuals with PTLD in this study had significantly longer follow-up time than those without PTLD (Table 1). To address this, Cox proportional hazards models were used to model time to the development of PTLD. As can be seen in Table 2, effect estimates were generally similar between the multivariable logistic model and the multivariable Cox proportional hazards model. To determine if year of transplant
TABLE 1.
Characteristics of controls, EBV-positive PTLD cases, and EBV-negative PTLD cases Descriptive characteristics
Controls
Age (mean ± SD), yr Male sex African American EBV-seronegative preransplantb Type of transplantc LRD kidney LUD kidney Deceased donor kidney Kidney + pancreasd Other—liver, heart, lung, pancreas aloned Reason for transplante Diabetes Hypertension Glomerulonephritis Study follow up time (mean ± SD), yr Detailed characteristics of PTLD cases Years from transplant to diagnosis (median, IQR) B-cell PTLD T-cell PTLD Documented CNS involvement Documented marrow involvement Lymphoma on pathology/monomorphic PTLD Graft failure after PTLD Removal of graft (nephrectomy or pancreatectomy) Median survival years after diagnosis
Pa
EBV-positive PTLD
EBV-negative PTLD
45.8 ± 14.9 60.1% (836/1392) 19.8% (275/1392) 6.38% (73/1144)
45.6 ± 14.6 76.5% (62/81) 8.64% (7/81) 40.0% (14/35)
47.5 ± 13.2 60.0% (15/25) 0% (0/25) 0% (0/6)
0.83 0.013 0.002
