Posttransplant lymphoproliferative disorder (PTLD) remains a significant complication in pediatric patients undergoing solid organ transplant (SOT). The majority involve Epstein-Barr virus (EBV)–driven CD20+ B-cell proliferations, which respond to reduction of immunosuppression and anti-CD20–directed immunotherapy. Owing to the low overall incidence, prospective studies of pediatric PTLD are scarce, leading to a lack of comprehensive understanding of this disorder in pediatric populations. This review aims to bridge this knowledge gap by providing a comprehensive analysis of the clinical, morphologic, and molecular genetic features of PTLD in children, adolescents, and young adults after SOT.
To examine the clinical features, pathogenesis, and classification of pediatric PTLDs after SOT.
Personal experiences and published works in PubMed.
PTLD includes a broad and heterogeneous spectrum of disorders, ranging from nonmalignant lymphoproliferations to lymphomas. While most pediatric PTLDs are EBV+, an increasing number of EBV− PTLDs have been recognized. The pathologic classification of PTLDs has evolved in recent decades, reflecting advancements in understanding the underlying pathobiology. Nevertheless, there remains a great need for further research to elucidate the biology, identify patients at higher risk for aggressive disease, and establish optimal treatment strategies for relapsed/refractory disease.
Solid organ transplant (SOT) is an established and potentially curative treatment for children with organ dysfunction or failure.1–3 Following SOT, patients usually require lifelong immunosuppressive medications to maintain the health of the graft. These immunosuppressive medications may cause side effects, including infection and potential neoplasm.1 Posttransplant lymphoproliferative disorder (PTLD), a clinically and pathologically heterogeneous group of lymphoproliferative disorders, accounts for approximately 70% of pediatric neoplasms following SOT.1,3–5 The mortality rate associated with PTLD is up to 50% in some studies.3 Nevertheless, multicenter studies of pediatric PTLD are scarce owing to its overall low incidence. This review focuses on the clinical, morphologic, and molecular genetic features of PTLD in children, adolescents, and young adults after SOT.
EPIDEMIOLOGY
Since it was first reported in 1968,6,7 PTLD has been encountered in both SOT and hematopoietic stem cell transplant (HSCT) with variable incidences. A recent national registry of pediatric SOT recipients in the United States from 2005 to 2014 reported a 5-year cumulative incidence of PTLD ranging from 2% to 15.8% in pediatric transplant recipients.8 The highest incidence occurs in lung, intestine, and multiorgan transplant, followed by heart, liver, kidney, and others.1,9–11
RISK FACTORS
Multiple risk factors contribute to PTLD, including Epstein-Barr virus (EBV) status at the time of transplant, transplanted organ, and immunosuppressive regimen. The risk of PTLD is lower in kidney transplant recipients than in heart and lung transplant recipients, which might be associated with the level of immunosuppression needed in the respective organs or could also be due to the amount of passenger lymphocytes present in the graft.1,12 Additional risk factors include the age and race of transplant recipients.13 The incidence of PTLD following SOT appears relatively high in White patients younger than 10 years.1,12 Nevertheless, the EBV serostatus of donor or recipient is one of the most important risk factors for PTLD development.13–15 The incidence of PTLD is highest among EBV-seronegative recipients, who are at risk for primary EBV infection following transplant.15 EBV surveillance by polymerase chain reaction (PCR) is part of routine evaluations of transplant patients in many transplant centers. Most pediatric PTLDs present with EBV DNAemia.16 Generally, higher EBV DNA level is associated with higher risk of PTLD, although there is no arbitrary cutoff value to better stratify the risk and prognosis of PTLD.17 Moreover, since EBV PCRs are not standardized across centers, each center has developed its own cutoff, and results from different laboratories cannot be compared.
PATHOGENESIS
The main driver of PTLD pathogenesis is the combination of impaired immunity and EBV infection.1 More than 90% of pediatric PTLD cases are EBV positive, and the oncogenic effect of EBV has been well studied.18 In mouse models, latent membrane protein (LMP1) 1 acts as a classic oncogene.
EBV is a double-stranded DNA virus in the gamma herpesvirus family with tropism for B cells; once infected, B cells can serve as a lifelong reservoir. EBV seroprevalence increases by age, from 54% in children aged 6 to 8 years to 82.9% in young adults aged 18 to 19 years to more than 95% in patients older than 35 years. According to viral protein expression and proliferative and infectious capabilities, B cells can be in a lytic phase or 4 distinct latency phases. In latency 0, the viral genome persists within B cells without any expression of viral proteins. Only EBV-encoded small RNAs (EBERs) are expressed. EBV is thus not recognized by the immune system and can persist lifelong in the B cell. In latency I, only Epstein-Barr nuclear antigen (EBNA) 1 is expressed. Latency II is characterized by LMP1 and LMP2 expression in addition to EBNA1. EBV-associated malignancies in immunocompetent hosts are usually associated with latency I or II; for example, Burkitt lymphoma (BL) is associated with latency type I and Hodgkin lymphoma is associated with latency II. In latency III, all 6 EBNA proteins and 3 LMP proteins are expressed, thus making this a highly immunogenic state. This latency type is more commonly encountered in malignancies associated with immunodeficiency such as PTLD and AIDS-associated lymphomas.19,20 In the immunocompetent host, infection with EBV is countered by a cytotoxic T-lymphocyte (CTL) response that controls the infection, and EBV persists in B cells in latency type 0. Proliferation requires reexpression of viral antigens that restimulate the CTL response. After SOT, patients lack the ability to mount a robust CTL response owing to the immunosuppression needed to tolerate the graft and are thus at increased risk for uncontrolled B-cell lymphoproliferation or PTLD.
EBV-negative PTLD tends to occur late after transplant, and is usually monomorphic type, with a median onset of 6.6 years as compared to 3.3 years for EBV-positive PTLD.21,22 The pathogenesis of EBV-negative PTLD also involves suppression of antiviral immune activity, impaired immune surveillance of tumor cells, and dysregulation of molecular signaling/DNA repair mechanisms.1 Secondary genetic abnormality of EBV-negative PTLD is assumed to be different from that of EBV-positive PTLD, although more studies in large patient cohorts are warranted.23,24
PATHOLOGY CLASSIFICATION
In the 2017 World Health Organization (WHO) classification, PTLD was listed as 1 of 4 immunodeficiency-associated lymphoproliferative disorders, which was based on combined information of transplant history, histology, immunophenotype, and genetic diagnostic criteria.25 The newly released WHO classification and International Consensus Classification (ICC) have updated terminology and diagnostic criteria, as summarized in Table 1.26,27
Nondestructive Posttransplant Lymphoproliferative Disorder
Nondestructive posttransplant lymphoproliferative disorder (ND-PTLD) is more commonly seen in children with solid organ transplant than adults and often affects adenoids, tonsils, lymph nodes, and gastrointestinal tracts. The 2017 WHO classification acknowledges 3 histopathologic subtypes of ND-PTLD: plasmacytic hyperplasia (PH), infectious mononucleosis–like (IM-like), and florid follicular hyperplasia (FFH).4,28 The affected tissue architecture should be intact in ND-PTLD. In addition, PH is characterized by the predominance of plasma cells expanding the medullary cords and occasionally extending to the interfollicular areas. These plasma cells should be polytypic and negative for B-cell receptor gene rearrangements. The diagnosis of PH type of ND-PTLD relies on the presence of EBV-positive plasma cells.29 IM-like PTLD is characterized by features of infectious mononucleosis, with paracortical expansion by an immunoblast-rich infiltrate within a mixed background of T cells and plasma cells. Similarly, the histologic features of FFH type of ND-PTLD cannot be distinguished from FFH in immunocompetent patients if there is absence of EBV within follicles. Importantly, scattered EBV-positive cells can be seen in lymphoid or epithelial cells in transplant recipients,4,18 which does not warrant a diagnosis of ND-PTLD, although there are no established criteria for distribution and number of EBV-positive cells to differentiate them from ND-PTLDs thus far.29
It is recommended to recognize ND-PTLD in patients with a suspicious condition because a subset of ND-PTLDs have been associated with concurrent or subsequent polymorphic or monomorphic PTLDs. Genetic studies have revealed clonal cytogenetic abnormalities in ND-PTLDs.24,28,30 For instance, recurrent mutations of NOTCH1, CREBBP, and BCL11B have been reported in ND-PTLDs.24 Nevertheless, some of these genetic abnormalities show low variant allele fraction and may originate from background cells, which can be seen in nonspecific reactive lesions.31 Currently, there is no solid evidence to prove a linear clonal evolution from ND-PTLD to polymorphic or monomorphic PTLDs.19,24
Because there is no cutoff value for EBER-positive cells or EBV viral load in the current diagnostic criteria of ND-PTLD, it is extremely challenging for pathologists to differentiate EBV infection/reactivation from ND-PTLD.4,32 For example, a patient who had received a heart transplant 9 years previously had a peripheral blood EBV DNA viral load less than 1000 copies/µL. Gastrointestinal endoscopy identified patchy erythema but absence of ulceration or mass lesions. Multiple sites were biopsied, which revealed intact tissue architecture with variable number of EBER-positive cells, ranging from rare to more than 100 per high-power field. The overall findings suggest an FFH type of ND-PTLD (Table 2) from the colon biopsy (Figure 1, A through F), with the context of heart transplant history and immunosuppression. The patient’s symptoms resolved with reduction of immunosuppressive (IS) therapy and 4 doses of rituximab. This case emphasizes the difficulty of differentiating ND-PTLD from EBV reactivation. In patients with suspected PTLD, occasionally multiple biopsies are required for an accurate diagnosis.
Frequency and Subtypes of Posttransplant Lymphoproliferative Disorder (PTLD) Following Pediatric Solid Organ Transplant

Colon, esophagus, and intestine biopsy specimens from a patient post heart transplant with diarrhea. The colon biopsy specimen shows an intact architecture with occasional lymphoid follicles (A, C, and D). Epstein-Barr–encoding region (EBER) in situ hybridization (ISH) identified an increase in EBER+ cells within the follicle in the colon biopsy specimen (B), suggestive of a follicular hyperplasia variant of nondestructive posttransplant lymphoproliferative disorder (ND-PTLD). A differential diagnosis of infectious mononucleosis–like ND-PTLD was also considered. However, only scattered EBER+ cells are seen in the esophagus biopsy specimen (E), with absence of EBER+ cells in the small-intestine biopsy specimen (F) (hematoxylin-eosin, original magnification ×200 [A]; CD20, original magnification ×200 [C]; CD3, original magnification ×200 [D]; EBER ISH, original magnification ×200 [B, E, and F]).
Colon, esophagus, and intestine biopsy specimens from a patient post heart transplant with diarrhea. The colon biopsy specimen shows an intact architecture with occasional lymphoid follicles (A, C, and D). Epstein-Barr–encoding region (EBER) in situ hybridization (ISH) identified an increase in EBER+ cells within the follicle in the colon biopsy specimen (B), suggestive of a follicular hyperplasia variant of nondestructive posttransplant lymphoproliferative disorder (ND-PTLD). A differential diagnosis of infectious mononucleosis–like ND-PTLD was also considered. However, only scattered EBER+ cells are seen in the esophagus biopsy specimen (E), with absence of EBER+ cells in the small-intestine biopsy specimen (F) (hematoxylin-eosin, original magnification ×200 [A]; CD20, original magnification ×200 [C]; CD3, original magnification ×200 [D]; EBER ISH, original magnification ×200 [B, E, and F]).
Polymorphic Posttransplant Lymphoproliferative Disorder
Polymorphic PTLD shows architectural destruction and is histologically characterized by an extensive polymorphic infiltrate, composed of a mixture of large lymphoid cells (such as immunoblasts), plasma cells, and small and medium-sized mature lymphocytes, occasionally accompanied by Hodgkin/Reed Sternberg-like (HRS-like) cells and necrosis. The HRS-like cells are often positive for CD30, CD20, CD79a, and PAX5, with occasionally weak CD15 expression. Most polymorphic PTLDs are positive for EBER in situ hybridization.29 Although EBV is the main driver of polymorphic PTLDs, additional genomic abnormalities can be identified in up to 30% of polymorphic PTLDs.20 Monoclonal immunoglobulin and/or T-cell clonality gene rearrangements have been discribed.33 Given the morphologic and immunophenotypic spectrum of polymorphic PTLD, defining the boundary of this entity with monomorphic PTLD is often challenging.20,29,34 For example, a subset of EBV-positive diffuse large B-cell lymphomas (DLBCLs) have a low density of neoplastic cells with or without HRS-like features.34,35 Moreover, it is challenging to differentiate polymorphic PTLD from 2 indolent variants of PTLD, including IM-like ND-PTLD and EBV-positive mucocutaneous ulcer (EBV+MCU) (Figure 2, A through F).
A palate ulcer biopsy specimen from a patient post heart transplant. The hematoxylin-eosin section shows a polymorphic infiltrate with occasional large lymphoid cells and numerous mature lymphoplasmacytic and inflammatory cells (A and D). PAX5 highlights B lymphocytes in 2 representative areas (B and E). CD3 stains T lymphocytes in the background (not shown). EBER ISH is positive in B cells with a cell size spectrum (C and F), consistent with a diagnosis of polymorphic posttransplant lymphoproliferative disorder, in the context of positive EBV DNAemia (original magnification ×400 [A and D]; original magnification ×200 [B, C, E, and F]). Abbreviations: EBER, Epstein-Barr–encoding region; EBV, Epstein-Barr virus; ISH, in situ hybridization.
A palate ulcer biopsy specimen from a patient post heart transplant. The hematoxylin-eosin section shows a polymorphic infiltrate with occasional large lymphoid cells and numerous mature lymphoplasmacytic and inflammatory cells (A and D). PAX5 highlights B lymphocytes in 2 representative areas (B and E). CD3 stains T lymphocytes in the background (not shown). EBER ISH is positive in B cells with a cell size spectrum (C and F), consistent with a diagnosis of polymorphic posttransplant lymphoproliferative disorder, in the context of positive EBV DNAemia (original magnification ×400 [A and D]; original magnification ×200 [B, C, E, and F]). Abbreviations: EBER, Epstein-Barr–encoding region; EBV, Epstein-Barr virus; ISH, in situ hybridization.
Patients with EBV+MCU present with well-circumscribed, often painful, ulcerating lesions restricted to mucosal or cutaneous sites without any mass lesions. EBV+MCU was first reported by Dojcinov et al,36 and it has recently been included in the spectrum of PTLDs. The morphologic features of EBV+MCU are characterized by a polymorphous infiltrate with atypical large lymphoid cells and occasional HRS-like cells.36 The base of the lesion is sharply defined by a rim of small T lymphocytes. The large lymphoid and HRS-like cells are EBV-positive B cells that are also positive for PAX5, MUM1, CD30, and CD20 with occasionally weak CD15 expression. The morphology and immunophenotype of EBV+MCU are nearly indistinguishable from those of polymorphic PTLD.29 A prior study of 7 patients with EBV+MCU within a cohort of 70 EBV+ PTLDs showed that none of the patients with EBV+MCU had EBV DNA in their blood (<1000 copies/µL) at diagnosis or follow-up,37 and all the patients had response to reduction in immunosuppression.36,37 Therefore, the absence of EBV viremia and solitary lesion are essential to differentiate EBV+MCU from polymorphic PTLD.20
Monomorphic Posttransplant Lymphoproliferative Disorder
Monomorphic PTLD is defined as lymphoid or plasmacytic proliferations that fulfill the diagnostic criteria according to the WHO classification system of non–transplant-related lymphomas. During the last decades, the spectrum of monomorphic PTLDs has been expanding and currently includes indolent B-cell lymphoma, aggressive B-cell lymphoma, classic Hodgkin lymphoma, T- and NK-cell neoplasm, and plasma cell neoplasm, all of which can be EBV positive or negative.20,23,26 Below we have summarized recent progress on each lymphoma subtype.
Aggressive B-cell lymphomas and aggressive B-cell lymphomas after SOT are often but not always EBV+ and include DLBCL; high-grade B-cell lymphoma, not otherwise specified (NOS); BL; and plasmablastic lymphoma.20 The most common subtype of monomorphic PTLD is DLBCL, comprising more than 60% of cases, often with high disease stage (stage III or IV) and elevated lactate dehydrogenase levels.1,38 The morphology of EBV-positive DLBCLs is usually characterized by a diffuse infiltrate of large lymphoma cells, although rare EBV-positive DLBCLs may show a focal increase of large lymphoma cells and a polymorphic background of reactive components such as small lymphocytes, plasma cells, histiocytes, and epithelioid cells. It can create a diagnostic challenge to differentiate from polymorphic PTLD in the small core biopsy. The prognostic significance of these 2 morphology variants is unclear. Lymphoma cells usually express CD19, CD20, CD22, and CD79 and more commonly show an activated B-cell subtype. CD30 shows positivity in a subset of cases. EBV shows positivity in up to 60% to 70% of DLBCL-PTLDs, with latency type II (LMP1 positive) or latency type III (EBNA2 positive).25
Burkitt monomorphic posttransplant lymphoproliferative disorder (BL-PTLD) is uncommon. The diagnostic criteria of BL-PTLD are similar to those for BL in immunocompetent patients. Recent studies of 20 BL-PTLDs showed that the median time from transplant to BL-PTLD was 7.2 years and approximately 70% of cases were EBV positive. Bone marrow involvement appeared to be associated with adverse prognosis. Recently, Salmeron-Villalobos et al39 investigated the genetic landscape of 31 pediatric monomorphic PTLDs after SOT, including 24 DLBCLs and 7 BLs, predominantly EBV positive. Interestingly, BL-PTLD carried mutations in MYC, ID3, DDX3X, ARID1A, or CCND3, resembling the genetic landscape of BL in immunocompetent patients, although fewer copy number alterations were identified in BL-PTLD.39 In contrast, DLBCL-PTLD showed a very heterogeneous genomic profile with fewer mutations and copy number alterations than DLBCL in immunocompetent patients. Recurrent epigenetic modifiers and genes of the Notch pathway were identified in approximately 28% of DLBCL-PTLDs, which might be associated with a worse outcome.39 This study also highlights the low genetic complexity of pediatric DLBCL-PTLD and its good response to low-intensity treatments.39
Central nervous system (CNS) involvement by PTLD is rare, with a variably reported interval from transplant to diagnosis of 1 year to 4 years. Most patients have multifocal disease and most cases are DLBCL-PTLDs. The optimal treatment of CNS-PTLD remains unknown.40–42 Based on recent studies,43,44 Burkitt-like lymphoma with 11q aberration, a germinal center–derived lymphoma, appears to be closer to high-grade B-cell lymphoma or diffuse large B-cell lymphoma than to BL. Plasmablastic lymphoma (PBL) and high-grade B-cell lymphoma–NOS are aggressive B-cell lymphomas occurring predominantly in adult patients and have very rarely been reported in pediatric patients.45–47 Occasionally, isolated pleural-based monomorphic-type PTLD has been reported, which is often EBER positive, HHV8 (human herpesvirus-8) negative, and diffuse large B-cell type with or without plasmacytic differentiation. Flow cytometry is helpful to detect monoclonal B cells in these unusual cases. Some patients had response to rituximab treatment, but not all.48–50
In 2011, Gibson et al51 reported 4 EBV-positive extranodal marginal zone lymphomas of mucosa-associated lymphoid tissue (MALTs) with plasmacytic differentiation in pediatric and adult SOT patients, and all 4 patients achieved complete remission after reduction of IS therapy, local surgical excision, rituximab, or local radiation therapy. The morphology and immunophenotype of these 4 EBV-positive MALTs were similar to EBV-negative MALT in immunocompetent patients. Genomic studies on EBV-positive MALTs revealed pathogenic/likely pathogenic mutations in IRF8, BRAF, TNFAIP3, and SMARCA4 and recurrent copy number abnormalities involving the IRF family or interacting genes (IRF2BP2, IRF2, and IRF4), whereas trisomies of chromosomes 3 or 18 and recurrent mutations in the NF-κB pathway (commonly seen in MALT of immunocompetent patients) were not identified.52 These findings indicate a different transforming mechanism for EBV+ MALT in transplant receipients.25 Recently, Galera et al53 expanded the spectrum of monomorphic PTLDs and reported clinical presentations and anatomic locations of EBV-negative marginal zone lymphomas (MZLs) in 9 SOT recipients, including 7 extranodal MZLs with gastrointestinal involvement and 2 nodal MZLs. Next-generation sequencing identified recurrent mutations previously reported in MZL, namely TNFAIP3, TNFRSF14, LRP1B, FAS, and NOTCH2.25,53 These EBV-positive or EBV-negative MZLs are diagnosed at least 1 year after SOT, and occur in both children and adults with no sex bias, often showing an indolent clinical course and responding to reduction of IS therapy, rituximab, or local therapy.53
Sporadic cases of other subtypes and rare locations of indolent B-cell lymphoma after SOT have been reported in the literature. For instance, an 18-year-old woman with history of heart transplant developed a breast mass consistent with MZL with massive amyloid deposition, which responded to reduction of IS therapy.54 A case of hairy cell leukemia was reported after cardiac transplant; it is unclear whether this was truly related to immunosuppression and therefore part of the spectrum of PTLD, or if it represents a coincidental event in this patient.55
Classic Hodgkin lymphoma (CHL) is not a distinct subtype of PTLD in the WHO 5th edition but remains a rare subtype of PTLD in the WHO 4th edition and ICC classification.26,27,56 According to a recent survey study of 192 CHL-PTLDs from the Scientific Registry of Transplant Recipients (SRTR), patients with CHL-PTLD tend to be older, more likely male and White, and have extranodal disease. Median time from transplant to CHL-PTLD diagnosis was 88 months, but stage and tumor EBV status were insufficient for analysis. Patients receiving CHL-like chemotherapy regimens had significantly improved survival.57 Although CHL has been included in the classification of PTLD, because of its low incidence, not much is known about the biologic, genetic, and clinical features. Further study of this interesting and uncommon subtype of PTLD is warranted. In contrast, PTLDs with HRS-like cells and preserved B-cell immunophenotype show different genetics, clinical course, treatment response, and prognosis from CHL-PTLD, and should not be considered as CHL-PTLDs.58,59 These cases usually are classified either as polymorphic PTLD or DLBCL type of monomorphic PTLD. For instance, Ranganathan et al60 reported on 70 young patients with PTLD harboring HRS-like cells with strong B-cell antigen expression, which differed from true CHL-PTLD highlighted by positive CD30, CD15, and EBER expression and negative CD20 and CD79a expression in HRS cells. The authors also demonstrated a unique instance of evolution from a PTLD with HRS-like cells to CHL-PTLD.60 Patients with prior metachronous non–CHL-PTLD were younger at time of transplant, had a longer latency time to CHL-PTLD, and had prolonged high-level EBV DNAemia.61
T- and NK-cell PTLD (T/NK-PTLD) is rare, representing 2% to 15% of all PTLD cases, which usually present late, with a median posttransplant interval of 6 years, and are associated with a poor prognosis.62,63 In contrast to B-cell PTLDs, only 30% to 40% of T/NK-PTLDs are EBV related.64 Peripheral T-cell lymphoma, NOS is the most common subtype of T/NK-PTLD, followed by anaplastic large cell lymphoma (ALCL), hepatosplenic T-cell lymphoma (HSTCL), monomorphic epitheliotropic intestinal T-cell lymphoma, adult T-cell lymphoma/leukemia, and other rare types. The diagnostic criteria of T-cell lymphoma are similar to those for immunocompetent patients. However, it is unclear whether most or only a small subset of T-cell lymphomas after SOT are etiologically linked to the immunosuppression or whether it represents a coincidental occurrence in the posttransplant setting.65 For instance, Rajakariar et al66 reported 4 cases of T-cell lymphoma (including HSTCL and ALCL) following renal transplant; in each case, patients presented with nonspecific symptoms, including pancytopenia and/or liver dysfunction, with no obvious lymphadenopathy, and eventually died of the disease.
It is also unclear whether there are “early” nondestructive T- and/or NK-cell lymphoproliferations after SOT, although various polyclonal or monoclonal T-cell proliferations are often seen after SOT, which occasionally have been overdiagnosed as T-cell lymphoma. For example, aberrant T-cell populations with or without monoclonality are found in EBV-associated hemophagocytic lymphohistiocytosis, which does not necessarily indicate an underlying T-cell lymphoma.65,67,68 Thus far, most of these aberrant T-cell proliferations (such as large granular lymphocytosis) not filling the criteria of the abovementioned T-cell lymphoma have not been included in the diagnostic spectrum of PTLD.
EBV-negative T/NK-PTLDs also occur after SOT, but survival appears to be inferior than for EBV-positive cases. However, this finding is inconclusive, mainly because most of the current literature reports contained only single cases, with very few having more than 3 cases.63 A large cohort study of 17 adult T/NK-PTLDs revealed recurrent mutations of epigenetic modifier genes (TET2, KMT2C, KMT2D, DNMT3A, ARID1B, ARID2, KDM6B) in 11 cases, inactivation of TP53 by mutation and/or deletion in 6 cases, and mutations of JAK/STAT pathway genes in 5 cases. Complex copy number changes were detected in 50% of cases. The authors reported that the molecular and genomic alterations observed in T/NK-PTLDs appear similar to those reported for mature T-cell lymphomas in immunocompetent hosts.69 However, this study only included rare EBV-positive T/NK-PTLDs, probably insufficient to assess the difference between EBV-positive and EBV-negative T/NK-PTLDs. Nevertheless, there is an absence of genetic studies on pediatric T/NK-PTLD, mostly owing to the limited number of cases.
Plasma cell myeloma and plasmacytoma-like PTLDs represent 3% to 6% of PTLDs, with a median interval of 4.8 years from SOT to PTLD. A survey study of 212 patients with multiple myeloma–PTLD from SRTR revealed a male predominance, frequent extramedullary disease, and advanced age at diagnosis, with elevated creatine level (>2 g/dL), White race, and use of the monoclonal antibody muromonab CD3 (OKT3) all associated with poor prognosis.70 EBV status of the tumor was not available for analysis in this study.70 Nevertheless, posttransplant-associated plasma cell neoplasms are rare in pediatric patients, with only rare cases reported in the literature.71 Additional cases of pediatric plasmacytoma-like PTLD have been reported in heart, renal, or combined small bowel–liver transplant recipients.72,73
It is well known that some of these plasmacytomas are EBV positive.47 Importantly, EBV-positive plasmacytoma likely represents a variant of plasmacytoma (Figure 3, A through D). A recent study by Zhou et al47 showed that the survival rate of patients with an EBV-positive plasmacytoma was comparable to that of patients with EBV-negative plasmacytoma. On histology, EBV-positive plasmacytoma is less commonly associated with a “starry-sky” appearance, necrosis, absence of light-chain expression, and a high Ki-67 index (>75%), in contrast to PBL. Genetic features of EBV+ plasmacytoma are also different from PBL.47 Similarly, it is challenging to separate plasma cell myeloma with plasmablastic morphology from PBL,74,75 while the treatments of these 2 diseases are quite different.
An EBV+ plasmacytoma in a young child after transplant. The hematoxylin-eosin section shows a diffuse infiltrate of mature plasma cells (A), positive for CD138 (B) and EBER ISH (D) but negative for CD20 (C). The plasma cells are light-chain restricted (not shown) (original magnification ×400 [A]; original magnification ×200 [B, C, and D]). Abbreviations: EBER, Epstein-Barr–encoding region; EBV, Epstein-Barr virus; ISH, in situ hybridization.
An EBV+ plasmacytoma in a young child after transplant. The hematoxylin-eosin section shows a diffuse infiltrate of mature plasma cells (A), positive for CD138 (B) and EBER ISH (D) but negative for CD20 (C). The plasma cells are light-chain restricted (not shown) (original magnification ×400 [A]; original magnification ×200 [B, C, and D]). Abbreviations: EBER, Epstein-Barr–encoding region; EBV, Epstein-Barr virus; ISH, in situ hybridization.
Synchronous and Metachronous Posttransplant Lymphoproliferative Disorder
It is important to recognize that nondestructive, polymorphic, and monomorphic PTLDs can occur at the same time or sequentially in the same patient.76,77 For example, Figure 4, A through H, shows the case of a patient with ND-PTLD in the lymph node and a synchronous DLBCL-PTLD in the adjacent extranodal soft tissue. This patient achieved remission after CHOP (cyclophosphamide + doxorubicin + vincristine + prednisone)–based chemotherapy. However, a year later, another new tonsil lesion consisting of polymorphic PTLD developed (figure not shown), and the patient eventually died of disease progression. Similarly, another patient presented with a palate ulceration 9 years after heart transplant. A polymorphic PTLD was diagnosed (Figure 2, A through F), and the patient received rituximab. One week later, the patient presented with a small-intestine perforation, which was involved by MZL-like monomorphic PTLD (Figure 5, A through E). PCR clonality assessment of the palate and intestine lesions revealed 2 distinct B-cell clones. Similarly, Mandell et al78 reported on a liver transplant patient who developed a clonally distinct EBV-positive CNS DLBCL-PTLD, which was morphologically and immunohistochemically indistinguishable from the patient’s intra-abdominal DLBCL, diagnosed 5 months previously. Additional case reports further confirm the distinct clonality in synchronous and/or metachronous PTLDs at different sites.79
A boy presented with enlarged lymph node and hilar mass. A nondestructive PTLD was identified in the lymph node excision biopsy, which revealed polyclonal plasma cell hyperplasia (A) with scattered EBER positivity (B). κ (C) and λ (D) ISH are polyclonal. In the hilar mass, extensive necrosis and diffuse large B-cell lymphoma were identified on hematoxylin-eosin sections (E). The DLBCL cells are EBER positive (F) with λ restriction (H and G) (hematoxylin-eosin, original magnification ×400 [A and E]; EBER ISH, original magnification ×200 [B and F]; κ ISH, original magnification ×200 [C and G]; λ ISH, original magnification ×200 [D and H]). Abbreviations: DLBCL, diffuse large B-cell lymphoma; EBER, Epstein-Barr–encoding region; ISH, in situ hybridization; PTLD, posttransplant lymphoproliferative disorder.
A boy presented with enlarged lymph node and hilar mass. A nondestructive PTLD was identified in the lymph node excision biopsy, which revealed polyclonal plasma cell hyperplasia (A) with scattered EBER positivity (B). κ (C) and λ (D) ISH are polyclonal. In the hilar mass, extensive necrosis and diffuse large B-cell lymphoma were identified on hematoxylin-eosin sections (E). The DLBCL cells are EBER positive (F) with λ restriction (H and G) (hematoxylin-eosin, original magnification ×400 [A and E]; EBER ISH, original magnification ×200 [B and F]; κ ISH, original magnification ×200 [C and G]; λ ISH, original magnification ×200 [D and H]). Abbreviations: DLBCL, diffuse large B-cell lymphoma; EBER, Epstein-Barr–encoding region; ISH, in situ hybridization; PTLD, posttransplant lymphoproliferative disorder.
The same patient as in Figure 2. After diagnosis of polymorphic posttransplant lymphoproliferative disease in a palate ulcer biopsy, the patient received rituximab treatment and presented with a small-intestine perforation a week later. The hematoxylin-eosin section (A) of the perforation site shows numerous mature plasma cells, which are EBER+ (B), and λ restricted (D). κ Light-chain immunostaining is negative (E). Plasmacytoma was considered. However, scattered PAX5+ B cells with a size spectrum are present in the background (C), which favored extranodal marginal zone B-cell lymphoma. The polymerase chain reaction for B-cell clonality showed a distinct B-cell clone from the palate ulcer biopsy specimen (not shown) (original magnification ×400 [A and B]; original magnification ×200 [C, D, and E]). Abbreviation: EBER, Epstein-Barr–encoding region.
The same patient as in Figure 2. After diagnosis of polymorphic posttransplant lymphoproliferative disease in a palate ulcer biopsy, the patient received rituximab treatment and presented with a small-intestine perforation a week later. The hematoxylin-eosin section (A) of the perforation site shows numerous mature plasma cells, which are EBER+ (B), and λ restricted (D). κ Light-chain immunostaining is negative (E). Plasmacytoma was considered. However, scattered PAX5+ B cells with a size spectrum are present in the background (C), which favored extranodal marginal zone B-cell lymphoma. The polymerase chain reaction for B-cell clonality showed a distinct B-cell clone from the palate ulcer biopsy specimen (not shown) (original magnification ×400 [A and B]; original magnification ×200 [C, D, and E]). Abbreviation: EBER, Epstein-Barr–encoding region.
Rare cases of both B-cell and T-cell PTLDs in the same patient have been published. For instance, Morovic et al80 reported that a patient who had received a heart transplant 10 years previously developed 2 metachronous EBV-related PTLDs, including a polymorphic B-cell PTLD that completely regressed after reduction of IS therapy, and then a monomorphic T-cell PTLD after 31 months. The patient eventually died of disease progression with T-cell PTLD.80 In a study of 13 patients with CHL-PTLD, 6 had antecedent non–CHL-PTLD: 3 had polymorphic PTLD; 2, monomorphic PTLD; and 1, ND-PTLD. Patients with prior metachronous non–CHL-PTLD were younger at time of transplant and had a longer latency time to CHL-PTLD post transplant.61 Therefore, it is important to diagnose and follow up all variants of PTLD.
CLINICAL PRESENTATION
The clinical presentation of PTLD is highly variable, usually but not always depending on the location and histopathology. While transplant recipients may have low EBV viral loads during posttransplant surveillance, the marked elevation of EBV warrants a workup to rule out PTLD.81 In one study, plasma from patients with PTLD had a median EBV viral load of 3225 copies/100 μL; in comparison, the plasma from patients without PTLD had a median EBV viral load of 740 copies/100 μL.81 At early onset, PTLD may present with symptoms related to viral infection such as tonsilitis, sore throat, or tonsil hypertrophy, because EBV enters the body through the oropharyngeal epithelium and tonsils. The gastrointestinal tract is another common site in pediatric patients who can present with obstructive symptoms, gastrointestinal bleeding, diarrhea, and/or abdominal pain. Nonspecific constitutional symptoms such as fever, weight loss, and fatigue are common.
As disease advances, PTLD may present with organomegaly and masses involving the gastrointestinal tract (stomach, intestine), lungs, skin, liver, CNS, and the allograft itself,1,10,22 associated with organ- and location-related symptoms and/or laboratory abnormalities. Similar to nontransplant patients with lymphoproliferative disorders, imaging studies to stage the disease should be considered.1 Additional laboratory tests for PTLD include a comprehensive metabolic panel, lactate dehydrogenase, uric acid, and blood cell counts.1,10,22,82
THERAPIES
The treatment of pediatric patients with PTLD requires a multidisciplinary approach including oncology, transplant, and infectious disease teams. The first step is reduction of immunosuppression (RIS) to restore CTL function and provide an immune response. While at least a 50% reduction is recommended, this must be weighed against the risk of graft rejection and/or loss. RIS has shown success rates ranging from 43% to 63% and is generally thought to be more efficacious in ND-PTLD and polymorphic PTLD.1
Most pediatric PTLDs are EBV+ CD20+ B-cell proliferation disorders. In pediatric transplant patients with CD20+ EBV+ PTLD (polymorphic and monomorphic) that is nonresponsive to RIS, rituximab as monotherapy or in combination with chemotherapy is considered standard of care with overall response rates (ORRs) ranging from 44% to 69% (25%–53% complete response [CR]).1 In 38 pediatric heart transplant recipients with PTLD, a CR rate of 71% was reported with rituximab monotherapy.83 The PED-PTLD-200584 trial in Germany used a response-adapted approach with pediatric patients, with 3 weekly infusions of rituximab followed by response assessment. Patients with CR or partial response (PR) received an additional 3 weekly infusions of rituximab, while patients with stable or progressive disease received chemotherapy. Of 49 SOT patients, 32 achieved remission with rituximab alone, with only 6 relapses. These data show that responders to monotherapy with rituximab have good long-term outcomes.85
Chemotherapy for EBV+ PTLD is an acceptable first-line therapy in fulminant or advanced-stage monomorphic PTLD as well as salvage therapy after failed RIS and rituximab. The Children’s Oncology Group (COG) study ANHL0221 evaluated a combination of rituximab, prednisone, and low-dose cyclophosphamide for EBV+ CD20+ PTLD. Fifty-five patients were enrolled, and 69% achieved complete remission, with 2-year overall survival (OS) of 83% and event-free survival (EFS) of 71%. This regimen was based on a prior phase II study of prednisone with a low-dose cyclophosphamide after failed RIS with ORR of 83%, 2-year OS of 73%, and EFS of 69% in 36 evaluable patients. In the German PED-PTLD 2005 trial,84 patients who did not have response to initial rituximab therapy were treated with the moderate chemotherapy regimen, mCOMP (day 1: vincristine, prednisone, cyclophosphamide and day 15: methotrexate; repeated every 4 weeks for 6 cycles). Fifteen of the initial 49 patients had stable or progressive disease after rituximab and received treatment with mCOMP; of these, 4 had disease progression requiring more intensive chemotherapy. The 2-year OS and EFS were 86% and 67%, respectively, for the entire patient cohort including the rituximab responders. Notably, 6 of 7 patients with Burkitt histology required chemotherapy.
There is no standard approach to relapsed/refractory PTLD; salvage chemotherapy or treatment with EBV-specific cytotoxic T cells, as available, would be appropriate. EBV-specific cytotoxic T cells (EBV CTLs) can be readily generated from allogeneic, autologous, or third-party EBV+ donors with the hypothesis that they would restore the EBV-specific CTL response in the immunocompromised recipient.86 EBV CTLs have been used in specialized centers in the last 2 decades with promising results. The use of autologous EBV CTLs has been challenging because of ongoing immunosuppression and the time needed to manufacture these cells, raising interest in third-party EBV CTLs. Third-party EBV CTL use was reported by Haque et al86 in 2002; in their study, 3 of 5 patients who were unresponsive to conventional treatment achieved CR with a bank of frozen partially HLA-matched EBV CTLs. In a subsequent phase 2 multicenter trial, 33 patients with PTLD (31 SOT recipients) had ORR of 64%, with 79% of patients being alive at 6 months. Kazi et al87 described the long-term follow-up of 64 patients, including 28 HSCT recipients and 20 SOT recipients, using a third-party bank of HLA-matched EBV CTLs. In the SOT cohort, the ORR was 75% with a long-term survival of 60%. The largest third-party allogeneic EBV CTL bank including 330 distinct products was reported by Prockop et al.88 Thirty-three HSCT and 13 SOT patients with PTLD were treated, with 68% of HSCT and 54% of SOT recipients achieving CR or sustained PR, showing safety, feasibility, and promising response data. The COG trial ANHL1522 evaluated rituximab followed by third-party LMP-specific CTLs in newly diagnosed, relapsed and refractory pediatric SOT with EBV+ CD20+ PTLD that did not respond to rituximab monotherapy, with final results pending. Commercialization of EBV CTLs is actively being pursued in phase II and III trials evaluating tabelecleucel.
EBV-negative PTLD is treated according to the underlying pathology, but using the treatment regimen developed for immunocompetent individuals is often complicated by the reduced chemotherapy tolerance of transplant recipients owing to reduced graft function or reduced renal function from calcineurin toxicity. Plasma cell neoplasm of PTLD is very rare in pediatric patients, with only rare cases reported in literature. Epperly et al71 reported a case of plasma cell myeloma type of PTLD in a pediatric liver transplant recipient with cytogenetic abnormalities of 1q duplication and t(8;14). The patient achieved a 6-year remission after chemotherapy and subsequent autologous stem cell transplant, before dying of complications of repeated liver transplant. Additional cases of pediatric plasmacytoma-like PTLD have been reported in heart, renal, or combined small bowel–liver transplant recipients.76 Many of these patients had good response to chemotherapy based on multiple myeloma protocols.
There is no standard of care for patients with T- or NK-cell PTLD. Unfortunately, these patients have poor response to therapy and prognosis is poor. Various T-cell–directed chemotherapy regimens have been explored.
CONCLUSIONS
PTLDs represent a broad spectrum of lymphoid proliferations and a complication of chronic immunosuppression after transplant, related to EBV activation, resulting in proliferation of EBV-positive B cells in most cases. Progress has been made on the clinical, pathologic, and genetic features of pediatric PTLDs, but we still have a lot to learn about this disease to develop better therapies for patients. More importantly, owing to the low incidence of the disease, collaborative studies are warranted in the future.
References
Author notes
Cheng and Wistinghausen contributed equally to this manuscript.
Competing Interests
The authors have no relevant financial interest in the products or companies described in this article.