Post-transplant Lymphoproliferative Disorders (PTLDs) are a set of rare, but potentially severe complications, in solid organ (SOT) and bone marrow transplant (BMT) recipients. The PTLDs comprise a heterogenous morphological spectrum, ranging from benign polyclonal hyperplasia to aggressive lymphomas. In 2017, the World Health Organization (WHO) reclassified PTLD categories into four main groups (non-destructive PTLDs, polymorphic PTLDs, monomorphic PTLDs and classic Hodgkin lymphoma PTLDs).1 Most PTLDs are B-cell neoplasms (with diffuse large B-cell lymphomas [DLBCLs], representing over 80% of the cases), and only 5 to 10% are T cell/NK cell types or Hodgkin lymphoma type.2

The Epstein-Barr virus (EBV) plays a significant role in the pathogenesis of PTLDs (up to 70% of the cases are EBV positive), associated with chronic immunosuppression and T cell inhibition, however, about 30% of the patients are EBV negative.1,3

The PTLD incidence varies depending on immunosuppressive regimens, the type of transplanted organ (1 - 2% kidney, liver and bone marrow; 5 - 10% heart, lung or bowel) and age.4,5

Several series of patients with PTLD have been published, all of which present limitations due to heterogeneity in the patient population, incomplete data on the histopathological subtype and the low frequency of this entity. In our country, available data on this entity is even more limited.2,6-9

The primary aims of our study were to characterize clinical and pathological features of PTLDs, survival (global and by histological subgroups), cumulative incidence and clinical outcome according to the EBV status.

The PTLDs comprise a spectrum of lymphoid proliferations, ranging from polyclonal hyperplasia to extremely aggressive lymphomas. Our study describes a cohort of 46 patients with PTLD from a 20-year retrospective sweep in the database of a University Hospital in Buenos Aires.

Monomorphic PTLDs (58.7%) were the subtype most frequently observed, with DLBCL being the most common histology (19 cases; 41.3%).

Other subtypes included Burkitt lymphoma (n = 5; 10.8%), plasmablastic lymphoma (n = 1; 2.1%), cutaneous MALT lymphoma (n = 1; 2.1%) and peripheral T-cell lymphoma CD4+ NOS (n = 1; 2.1%). Thirty-two (69.6%) patients presented extranodal disease. Classical Hodgkin lymphoma-type PTLD was observed in 3 cases (6.5%). Polymorphic PTLD represented 14 cases (30.4 %). Non-destructive early lesions were observed in 2 cases (4.3%).

The spread of histological subtypes is in agreement with data from other series. Yoon et. al. reported a higher incidence of non-destructive lesions (14%), while the distribution of polymorphic (30.2%) and monomorphic (51.2%) subtypes was similar.2 Similar to other series, the EBV+ PTLD was approximately 70%.12-14

The PTLD cumulative incidence was 1.84% (95%CI 1.77 - 1.91) for solid organ transplantation and 0.84% (95%CI 0.48 - 1.2) for bone marrow. The incidence of kidney and liver transplant recipients was slightly higher than that of other series. Yoon et. al. reported an incidence of 0.78% and 0.77%, respectively. Patients in this study were mostly liver and kidney transplant recipients, although there were also 9% of heart and 14% of bone marrow transplants. The median interval between the transplant and PTLD diagnosis was 44.5 months, similar to our study.2

The study by Francis A. et. al., using the Australian and New Zealand Dialysis and Transplant Registry, showed a PTLD incidence of 2% for kidney transplant recipients. This increased incidence might be related to the extended follow-up of patients (for more than 50 years) and the better prognosis for kidney transplant recipients, when compared to other transplanted organs.15

The experience in Argentina is more limited. The incidence in our study is consistent with that of Mendizabal M. et. al., which included liver transplant recipients at various centers in South America. The cumulative incidence in this study was 1.7%, with the mean time of 39.7 (SD ± 35.2) months from the transplant to PTLD diagnosis.8 Fernández MC et. al. published a 13-year experience in a population of pediatric liver recipients. The incidence of PTLD was 5.7%, with an average time from the transplantation to PTLD diagnosis of 25 months (range 7 - 47 months).9

The literature data on the PTLD incidence in bone marrow transplant recipients is more heterogenous. In a retrospective series of more than four thousand allogeneic hematopoietic stem cell transplants, Styczynski J et. al. reported an incidence of PTLD of 3.22%, ranging from 1.2% for the matched-family donor and 11.24% for mismatched unrelated donor recipients. [7] On the other hand, Yoon et. al. published a series of 747 bone marrow transplant recipients with a PTLD incidence of 0.8%. Both studies, as well as ours, showed a high percentage of PTLD development within one year of diagnosis.2 However, our series included only one bone marrow transplant patient with PTLD, therefore it is difficult to compare populations with the findings of these authors.

Of note, our cohort lacks post-heart, lung and bowel transplant PTLD cases, the three subgroups with the highest incidence reported in the literature.4,5 This phenomenon might in part be explained by the poor prognosis for lung and bowel transplant patients in our country and the low indication for transplantation in these patients.

The current standard of care is reduction in immunosuppression as the first-line treatment, which might lead to cure in non-invasive and polymorphic PTLDs.16-18 For the remaining PTLD subtypes, there was only one clinical trial showing good results after sequential treatment with rituximab followed by chemotherapy.17 Reduction in immunosuppression was only seen in 50% of the patients. This may be due to the fact that part of the PTLDs had been diagnosed before the application of the current treatments and standards of care and that the immunosuppressive regimen is administered by the physician indicating transplantation, rather than by hematologists, in our country. Fortunately, reduction in immunosuppression has become widely accepted in recent years.

Survival rates in our analysis were consistent with those of international series. Bishnoi et. al. published a cohort of 141 patients with an overall survival of 15 years, with a median of 5 and 1.32 years in adult and elderly patients, respectively.19 Trape R et. al. reported a survival of 69% at 3 years and 65% at 5 years; however, disease-free survival was 77 months, remarkably higher than ours.17 Styczynski J. et. al. observed a 12-month survival of 69.2% (CI 61 - 76) in bone marrow transplant recipients.7 In our study, survival at 5 years was 46.2%, strikingly higher than that of Mendizabal M. et. al., which may be due to the fact we included all PTLD subtypes, with a high share of kidney transplant patients, who have a better overall prognosis.8 Finally, overall survival was not different in the subtype of EBER+ PTLD patients. Although our hypothesis is exploratory and the result was negative, the number of patients in our cohort is low and further studies including more patients are needed to confirm this finding.

This is the largest cohort of patients in Argentina and the first to determine the incidence of PTLDs in solid organ and bone marrow transplants, making available local data for better understanding our patient characteristics and prognosis. Moreover, it contributes valuable information in terms of the EBER status and its association with the overall survival.

Our study carries some limitations. Because of its retrospective nature, some clinical and follow-up data might be missing; however, the characteristics of our hospital's medical records minimize the loss of information. In addition, the prospective studies published to date are scarce because PTLD is an uncommon complication and, thus, our cohort has the best cost-effective design to analyze this disorder. On the other hand, our series had no PTLD cases in heart, lung or bowel transplant recipients and there was only one bone marrow transplant patient and no information could be obtained for these subpopulations.

It is important to consider that pathologists in our country have little access to EBER in situ hybridization (ISH) methods. Although the EBV status in all our samples could be determined by the EBER ISH, this technique—the gold standard for tissue analysis—tends to be replaced by immunohistochemistry (IHC) for EBV proteins (such as LMP1 and EBNA-1/2), which is not the methodology of reference because of the false-negative rate.20

On the other hand, several studies performed gene expression profiling, in which there seems to be a role for cytotoxic immune antiviral signaling in EBV+ vs. EBV- cases, which would explain the modification of the microenvironment in germinal center cells in the presence of EBV+.21,22 These studies detected an increase in the immunotolerance signal (PDL1, IDO1) in EBV+. As we have checkpoint inhibitor drugs, this approach may be useful also in the PTLD. Unfortunately, the increase in graft-versus-host disease with this type of drug makes it necessary to evaluate the safety for this type of patient.20

The PTLD incidence was similar to that of previous series and the EBER did not appear as a relevant factor in our patient survival. The immunological characterization of the tumor microenvironment in the PTLD by complex gene expression profiling and identification of target pathways can be of vital importance to stratify patients adequately and better design therapies in the future.