Volume 116, Issue 4 p. 863-870
Original Article
Free Access

Primary central nervous system post-transplantation lymphoproliferative disorder

An International Primary Central Nervous System Lymphoma Collaborative Group Report

Robert Cavaliere MD

Robert Cavaliere MD

Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, The Ohio State University, Columbus, Ohio

Search for more papers by this author
Gina Petroni PhD

Gina Petroni PhD

Department of Biostatistics and Epidemiology, University of Virginia, Charlottesville, Virginia

Search for more papers by this author
Maria B. Lopes MD

Maria B. Lopes MD

Department of Pathology, University of Virginia, Charlottesville, Virginia

Search for more papers by this author
David Schiff MD

Corresponding Author

David Schiff MD

Division of Neuro-Oncology, Department of Neurology, University of Virginia, Charlottesville, Virginia

Fax: (434) 982-4467

Division of Neuro-Oncology, Department of Neurology, University of Virginia, Charlottesville, VA 22909-0432===Search for more papers by this author
First published: 05 January 2010
Citations: 179

The International Primary Central Nervous System Lymphoma Collaborative Group: Brian P. O'Neill, MD; Scott R. Plotkin, MD; Gustaaf W. van Imhoff, MD; Edward A. Neuwelt, MD; Kristoph Jahnke, MD; Jeffrey J. Raizer, MD; J. Lee Villano, MD; Lauren E. Abrey, MD; Tracy T. Batchelor, MD; Gerald Illerhaus, MD; and Sandra J. Horning, MD.

Abstract

BACKGROUND:

Primary central nervous system (CNS) post-transplantation lymphoproliferative disorder (PCNS-PTLD) is a rare complication of solid organ transplantation. The objectives of this study were to define the clinical, radiologic, and pathologic features of this disease and to explore the impact of treatment on patient outcomes.

METHODS:

The authors reviewed the databases of participating institutions of the International Primary CNS Lymphoma Collaborative Group for cases of PCNS-PTLD. Thirty-four patients who had pathologically confirmed PCNS-PTLD without evidence of systemic PTLD were investigated retrospectively.

RESULTS:

The median time from transplantation to diagnosis of PCNS-PTLD was 4.4 years. Disease usually was multifocal and involved any location of the brain but was most common in the cerebral hemispheres, usually in the subcortical white matter or basal ganglia. Radiographically, all lesions enhanced either homogenously or in a ring-enhancing pattern. Cerebral biopsy was required to establish diagnosis in most patients. Most patients had monomorphic, Epstein-Barr virus (EBV)-positive disease of B-cell origin. Response rates were high regardless of treatment type, and the median survival was 47 months. Age was the only factor predictive of survival.

CONCLUSIONS:

The current study demonstrated that PCNS-PTLD is typically an EBV-induced B-cell lymphoma that is responsive to treatment with favorable survival in many patients. An aggressive approach to tissue confirmation of diagnosis and treatment with chemotherapy or radiotherapy should be strongly considered. Cancer 2010. © 2010 American Cancer Society

Post-transplantation lymphoproliferative disorder (PTLD) is a well recognized complication of solid organ transplantation. Its incidence varies, depending on the graft recipient's age, the type and intensity of immunosuppression, and the organ type transplanted. The incidence varies from 1% after renal transplantation to as high as 20% among small bowel recipients, and it is higher in children.1 Epstein-Barr virus (EBV) plays a causative role, and up to 90% of tumors are associated with the virus. The World Health Organization categorizes tumors histologically as early lesions, polymorphic lesions, and monomorphic lesions. Lesions are typically of B-cell origin.2 Extranodal involvement is common in PTLD, although the central nervous system (CNS) is an uncommon site of disease, particularly in isolation.

Schneck and Penn first reported primary CNS-PTLD (PCNS-PTLD) in 1970.3 Since then, in addition to individual patient reports, only 45 patients have been reported in 3 case series spanning 3 decades, which highlights the relative rarity of this condition.4-6 During this time, our understanding of hematopoietic malignancies, their treatment, and their interaction with the CNS have progressed. Consequently, novel therapies and treatment paradigms are available. Similarly, the science of transplantation has advanced, and we have a better understanding of immunosuppression and its consequences. In light of the rarity of PCNS-PTLD and the advances in the fields of neuro-oncology, transplantation, and hemato-oncology, we sought to evaluate the epidemiology and clinical features of PCNS-PTLD as well as the impact of current therapies on survival. We report 34 patients with PCNS-PTLD collected from participants of the International Primary CNS Lymphoma Collaborative Group, an international, multidisciplinary group focused on the study and treatment of CNS lymphoma.

MATERIALS AND METHODS

PCNS-PTLD was defined as a lymphoproliferative disorder of the CNS without evidence of systemic PTLD in patients who previously underwent solid organ transplantation. Histologic confirmation of PCNS-PTLD was required. Participating institutions collected data on case report forms. Investigators reported basic demographics, date and type of transplantation, type of immunosuppression used, incidence of rejection, and EBV status of the patient and the graft. Although supplemental evaluations, including lumbar puncture, systemic imaging, and ophthalmologic and bone marrow examination, were nonuniform, experts in CNS lymphoma concluded that all patients reported had isolated CNS disease. Centralized pathology review was not performed, although a single pathologist with expertise in CNS lymphomas reviewed pathology reports. Neuroimaging was not centrally reviewed. Local reviewing physicians recorded the number and location of lesions and the character of the enhancement. Responses were based on criteria described by Macdonald et al.7 Investigators reported on the use of chemotherapy, radiotherapy, rituximab, and antivirals and on adjustment of immunosuppression and corticosteroids. We also assessed radiographic response and patient survival. A proportional hazards model was used to investigate simultaneous effects of potential predictor variables on survival. Age was assessed as a continuous variable. No adjustments were made for multiple comparisons. Institutional review boards of participating institutions approved the study.

RESULTS

Thirty-four patients who fulfilled the definition of PCNS-PTLD were identified, including 8 patients who were reported previously in the literature.4 Patient demographics, transplantation data, and immunosuppression data are listed in Table 1. Twenty-eight patients who were evaluated for human immunodeficiency virus status were negative. One transplantation, 14 transplantations, and 19 transplantations were performed in the 1980s, 1990s, and 2000s, respectively. Table 2 summarizes presentations of PCNS-PTLD, and Table 3 summarizes magnetic resonance imaging (MRI) findings. Lesion location on MRI was specified in 30 patients (88%). Isolated leptomeningeal and ependymal enhancement occurred in only 3 patients (10%). Of those with parenchymal MRI lesions, 17 patients (61%) had multiple lesions. Supplementary diagnostic evaluations are recorded in Table 4. Diagnosis was established by cerebral tissue acquisition in 31 patients (24 patients, biopsy; 4 patients, subtotal resection; 3 patients, unknown extent of resection). Diagnosis was established by cerebrospinal fluid (CSF) analysis in 2 patients. Another patient was diagnosed at autopsy. Four patients, all of whom underwent biopsy (2 stereotactic, 2 open), had postoperative hemorrhages at the biopsy site (2 were fatal). EBV status of the graft and recipient was known in only 6 patients and 5 patients, respectively. Of the 27 tumors that could be classified according to World Health Organization criteria, 20 were monomorphous, 5 were polymorphous, 1 was a T-cell lymphoblastic lymphoma, and 1 was Hodgkin lymphoma. Only 2 of 26 patients (7.6%) who had data available were EBV-negative (a patient who had a polymorphous PTLD diagnosed 11.5 years after a transplantation and a patient who had a large B-cell lymphoma diagnosed 3.8 years after transplantation).

Table 1. Demographics and Transplantation Data of 34 Patients With Primary Central Nervous System Post-Transplantation Lymphoproliferative Disorder
Characteristic No. (%)/ Median [Range]
Sex
 Male 22
 Female 12
Median age at first transplantation [range], y 38 [5-63]
Median age at diagnosis of PCNS-PTLD [range], y 43 [6-75]
Time from most recent transplantation to PCNS-PTLD
 Median [range], y 4.4 [0.2-17.2]
 ≤1 y 12 (35); Median. 0.5 y
 >1 y 22 (65); Median 7.6 y
 >10 y 7 (21)
Total no. of transplantations 38
Type of transplantat
 Kidney 19
 Kidney and pancreas 6
 Liver 5
 Lung 2
 Kidney and heart 1
 Heart 1
 Intestinal 0
Maintenance immunosuppression, n = 29
 Corticosteroids+ 15
 Cyclosporin+ 16a
 Tacrolimus+ 11a
 Mycophenylate 16
 Azathioprine 7
 OKT3 3
Induction immunosuppression, n = 14
 ATG/OKT3 4
 Corticosteroids 9
 Cyclosporin 5
 Mycophenylate 5
 Tacrolimus 3
 Azathioprine 3
 Alemtuzumab 1
Graft rejection 16
  • PCNS-PTLD indicates primary central nervous system post-transplantation lymphoproliferative disorder; OKT3, muromonab-CD3; ATG, antithymocyte globulin.
  • a Used as a single agent in 1 patient.
Table 2. Presentation of Patients With Primary Central Nervous System Post-Transplantation Lymphoproliferative Disorder
Median duration of symptoms (range), wk 3 (0.5-26)
Median ECOG PS (range) 2 (0-4)
Symptoms, % of patients
 Headache 34
 Hemiparesis 33
 Ataxia 29
 Aphasia 20
 Other mental status change 36
 Seizures 26
  • ECOG PS indicates Eastern Cooperative Oncology Group performance status.
Table 3. Neuroimaging of Primary Central Nervous System Post-Transplantation Lymphoproliferative Disorder (33 Patients)
Finding Patients, %
Enhancement 97
 Homogenous 41
 Heterogeneous 56
 Ring pattern 29
Lesion location
 Lobar 85
 Basal ganglia 39
 Periventricular 36
 Callosal 14
 Meningeal/ependymal 40
 Infratentorial 33
 Spinal cord 10
 Contact with ventricle 10
Table 4. Supplementary Evaluations
Evaluation No. of Patients
Cerebrospinal fluid
 WBC count, ×109/L 16
  >5 10
  Median (range) 12 (0-187)
 Protein, g/L 17
  >50 10
  Median (range) 71 (17-149)
 Cytology 20
  Positive 1
  Negative 14
  Atypical 4
  Suspicious 1
 Flow cytometry 12
  Negative 10
  Positive 1
Slit lamp examination 12
 Positive 0
 Negative 12
Serum LDH 21
 Normal 16
 1-2×Normal 3
 >2×Normal 2
  • WBC indicates white blood cell; LDH, lactate dehydrogenase.

At the time of last follow-up, 14 patients were alive at a median follow-up of 23 months, and 19 patients had died (excluding the patient who was diagnosed at autopsy). Figure 1 illustrates median survival for the entire group (47 months). Six patients died within 1 month of diagnosis. Eight of 12 patients who were diagnosed within 1 year of their most recent transplantation died at a median of 0.4 months. Of the 19 patients who died, 6 did not receive any treatment (or had immunosuppressant medication withdrawal only). Four patients died of systemic complications in the absence of progressive disease (urosepsis, pneumonia, pulmonary embolism, and declining performance status necessitating hospice admission). Two patients died while receiving initial therapy. Only 3 patients had documented progressive disease at the time of death. No cause of death was reported for 4 patients.

Details are in the caption following the image

This chart illustrates overall survival (Kaplan-Meier method).

Patients received a variety of therapies, as expected in a retrospective series that spanned 25 years (Table 5). Twenty-two patients received radiotherapy, systemic chemotherapy, and rituximab either alone or in combination (Table 6). Although 8 patients received antivirals, they were distributed among the other treatment groups; thus, it was difficult to assess the impact of anti-EBV drugs on outcome. Similarly, immunosuppression was decreased in most patients; therefore, it was difficult to separate its impact on outcome from other interventions. Two patients, however, received decreased immunosuppression only; 1 remained alive at 89 months of follow-up and had a complete response on a follow-up MRI study; and the other patient survived for 7.4 months, although no follow-up imaging study was obtained. On multivariate analysis, only age was predictive of survival (P = .001). Performance status, chemotherapy, rituximab, radiotherapy, antivirals, and time from transplantation to PCNS-PTLD were not predictive of outcome (P>.05).

Table 5. Initial Treatment for Primary Central Nervous System Post-Transplantation Lymphoproliferative Disorder
Variable No. of Patients
Corticosteroid dosage
 Increased 17
 Stable 8
 Decreased 4
Immunosuppression
 Decreased 25
 Stable 6
 Unknown 3
Antivirals initiated 8
Systemic chemotherapy 10
 Methotrexate (combination  or single agent) 7
 Temozolomide 2
 CHOP 1
Intrathecal chemotherapy
 Yes 5
 No 17
 Unknown 12
Radiation therapy 8
  • CHOP indicates cyclophosphamide, doxorubicin, vincristine, and prednisone.
Table 6. Initial Treatment, Evaluable Patient Outcomes, and Radiographic Responses
Treatment Patient Deaths Patients Alive at Last Follow-Up Radiographic Responses
No. Median Survival (range), mo No. Median Follow-Up (range), mo
RT and other therapiesa 7 26.4 (2.1-145.4) 1 24.2 CR, 5; PR, 2; PD, 1
Systemic chemotherapy without RTb 3 7.5 (6.9-8.2) 3 78.2 (24.5-175.0) PR, 3; SD, 1; PD, 2
Rituximab without chemotherapy or RT 1 3.4 7 23.4 (4.0-55.6) CR, 6; PR, 2
Rituximab ± other therapiesc 5 13.5 (2.1-47.7) 9 23.6 (4.0-55.6) CR, 7; PR, 6; PD, 1
  • RT indicates radiotherapy; CR, complete response; PR, partial response; PD, progressive disease; SD, stable disease; ±, with or without.
  • a Five patients received RT alone, 2 patients received systemic chemotherapy and rituximab, and 1 patient received rituximab without systemic chemotherapy.
  • b Four patients received rituximab.
  • c In addition to rituximab, 3 patients received systemic chemotherapy without RT, 2 patients received systemic chemotherapy and RT, and 1 patient received RT without systemic chemotherapy.

Three patients underwent treatment for progressive disease. One patient received whole-brain radiotherapy and was alive 1.7 years after progression. Another patient received intra-arterial methotrexate and intravenous cyclophosphamide with blood-brain barrier disruption and was alive 14.3 years after progression. The third patient received temozolomide and rituximab and died 6 months thereafter. Four additional patients did not receive treatment at progression and died soon thereafter.

DISCUSSION

PTLD is the second most common malignancy after skin cancer among adult solid organ transplantation recipients. CNS involvement is rare, especially in isolation. A review of the Israel Penn International Transplant Tumor Registry indicated that 15% of patients with PTLD had CNS involvement; and, in half of those patients, the CNS was the only site of disease.8 To our knowledge, our case series represents the largest published series of PCNS-PTLD. Our method of case ascertainment did not allow us to determine a denominator; therefore, we were unable to calculate the incidence of PCNS-PTLD. Nonetheless, the reporting institutions are major neuro-oncology referral centers. Thus, the incidence of PCNS-PTLD is rare.

The time from transplantation to the diagnosis of PCNS-PTLD of 4.4 years in the current study was longer than that reported previously. In 3 previous series, this interval was 12.6 months, 18 months, and 20 months.4-6 Eight patients in our series developed PCNS-PTLD (23%) ≥10 years after transplantation. Although it is believed that such a latent onset is uncommon,4 this closely approximates the 16% of cases reported by Snanoudj et al. that occurred >10 years after transplantation.5 In agreement with other series, 35% of patients developed PCNS-PTLD within 1 year of transplantation.5, 6 Unfortunately, the number of patients was insufficient for a meaningful comparison of those who were diagnosed sooner or later after transplantation, although this issue requires additional examination.

The majority of patients had multifocal disease (61%) with a predilection for the periventricular/basal ganglia region (63%), in agreement with Snanoudj et al., who reported this in 72% and 72% of patients, respectively.5 Five of 8 patients reported by Phan et al. also had “ependymal contact.”4 Castellano-Sanchez et al. also noted that the majority of patients with PCNS-PTLD had multiple lesions (83%). Thirty-three percent of our patients had infratentorial involvement, also in agreement with Snanoudj et al. (24%). Lesions were enhanced with contrast in all of our patients with homogenous and ring-enhancement patterns in 41% and 29%, respectively. This differs from other series in which the majority of patients had ring-enhancing lesions.4, 5 The difference may be explained by our method of case ascertainment.

Brain lesions in immunosuppressed patients remain a diagnostic dilemma. The differential is broad and includes infectious (abscess, toxoplasmosis, etc) and neoplastic (PTLD and others) etiologies, the risk of which increases substantially in an immunosuppressed population. The decision to perform less invasive diagnostic evaluations or proceed immediately to biopsy is controversial.9, 10 Our data support the use of cerebral biopsy. Although the majority of our patients had abnormal CSF, the abnormalities were nonspecific. Only 1 patient (5%) and 2 patients (17%) had definitively abnormal cytology results and flow cytometry results, respectively. Similarly, ocular examination may be of low yield, because none of the 12 patients who had this examination had ocular disease. However, these data must be interpreted with caution given the small number of patients assessed.

Certain factors may increase the risk of PTLD. One factor is transplantation type, and the greatest incidence is observed in intestinal and lung transplantations. This may be related to the intensity of immunosuppression or the quantity of lymphoid tissue within the transplanted graft.1, 11 In our series, none of the cases of PCNS-PTLD occurred in intestinal transplantation patients, and only 2 occurred in lung transplantation patients. The majority occurred in kidney recipients (26 patients). Renal transplantations are more common and, thus, account for more cases despite the lower overall risk. Alternatively, in PTLD that occurs in intestinal or lung recipients, the graft itself is frequently the site of disease.12 Because our patients had isolated CNS disease, such cases would have been excluded.

The type and intensity of immunosuppression have been linked with PTLD risk. Specifically, cyclosporine and tacrolimusmay increase the risk of PTLD.1 Twenty-five of 29 patients for whom we had data received either of these 2 agents. With 1 exception, however, all patients had undergone their transplantation after 1990, when these agents were used routinely. Therefore, this finding may not be unexpected. Although we believed it carried a lower risk, 5 of our patients received mycophenolate mofetil; however, O'Neill et al. have suggested an increased risk of CNS-associated EBV lymphoproliferative disorder with this agent in patients with autoimmune disorders.13 Snanoudj et al. also noted that several of their PCNS-PTLD cases developed soon after immunosuppression regimens were changed to mycophenolate mofetil.5 The intensity of immunosuppression, possibly a more significant factor than the specific agent, was unavailable for review.

Histologically, most cases in our series were monomorphic, a more advanced and aggressive form of PTLD. In addition, the majority of disease was of B-cell origin. Similar results were reported in 2 other series of PCNS-PTLD.4, 6 Although a higher proportion of patients (79%) had polymorphous disease in the series reported by Snanoudj et al., they also were of B-cell lineage.5 The single case of a T-cell lymphoma is unique among the PCNS-PTLD series reported to date; however, it does occur in systemic PTLD, in which T-cell lymphomas account for 10% to 15% of cases1, 14

Primary infection of a previously EBV-seronegative patient during immunosuppression has been identified as a significant risk factor for the development of PTLD. This may occur through natural exposure or through transmission from an EBV-infected graft in a seronegative recipient.11 In adults, this may be less relevant, because most adults already have acquired the infection during adolescence. Therefore, the absence of data regarding the EBV status of the recipient or graft in our series is less significant, because only 2 of our patients were aged <20 years at the time of their transplantation.

The optimal treatment of PCNS-PTLD remains unknown. The clinical status of patients may be compromised at presentation. Consequently, the risks of treatment may be greater than the risks for immunocompetent patients with CNS lymphoma. Because immunosuppression is recognized as a causative factor for PTLD, reduction of immunosuppression is performed routinely. With systemic PTLD, this reportedly induced remission in as many as 50% of patients without any additional therapy.11, 15 This is particularly true for patients who have polymorphous PTLD. Among patients with acquired immunodeficiency syndrome (AIDS)-related CNS lymphoma, reconstitution of the immune system with the initiation of highly active antiretroviral therapy at or after diagnosis prolongs survival.16, 17 Case reports suggest that a similar approach may be effective in PCNS-PTLD, although this has not been evaluated extensively. In our series, few patients underwent reduction of immunosuppression with curative intent without receiving additional therapy. Two patients had prolonged survival with this approach, suggesting a role for immune reconstitution. Most patients received other therapies simultaneously; therefore, it is difficult to assess the impact of reduced immunosuppression. Decreased immunosuppression, moreover, may place the graft at risk. Consequently, treatment should be performed in conjunction with transplantation specialists.

Antiviral therapies, most commonly ganciclovir and acyclovir, also have been considered for EBV-driven malignancies. A major limitation is that these thymidine kinase inhibitors are capable of inhibiting lytic viral replication but have no effect on tumor cells because of the absence of thymidine kinase in latently infected B-cells. Efforts to increase the thymidine kinase gene expression, including arginine butyrate and radiotherapy, are ongoing.18, 19 Although antiviral agents were administered to several of our patients, treating physicians incorporated other therapies concomitantly, and it was not possible to evaluate their effect.

The radiosensitivity of lymphoma long has been recognized in both immunocompetent patients and immunosuppressed patients.20 Our series demonstrated a high response rate among patients who received radiotherapy. In addition, the median survival for patients who received radiotherapy with or without other concomitant treatments was 26 months. Snanoudj et al. also observed an improved survival (median, 36 months vs 7 months for patients who did not receive radiotherapy) and response rates (85%) among patients who received radiotherapy with or without chemotherapy.5 Thus, radiotherapy remains a viable option for patients with PCNS-PTLD. In addition, radiotherapy is associated with less systemic toxicity and, thus, may be a more viable option for patients who already are compromised. The retrospective nature of our series precluded accurate assessment of the degree of palliation or toxicity.

Chemotherapy is the standard treatment in immunocompetent patients with PCNS lymphoma. Not uncommonly, immunocompromised patients with lymphoma have concomitant derangements, which may preclude chemotherapy. A recent report of AIDS-related PCNS lymphoma, however, suggested that high-dose methotrexate regimens may be tolerated well and are associated with excellent radiographic responses.21 Similarly, methotrexate was tolerated well among pediatric patients with PCNS lymphoma-PTLD.22 In our series, 4 of 6 patients who received primary treatment with systemic chemotherapy had a favorable response. Although response rates appeared to be lower with chemotherapy than with radiotherapy, half of the chemotherapy-treated patients were alive at the >2-year follow-up. Unfortunately, the retrospective nature of the study precluded an accurate assessment of toxicity. Nonetheless, these findings suggest that chemotherapy should be considered for patients with PCNS-PTLD in the absence of a systemic contraindication.

In phase 2 studies, single-agent rituximab, a humanized anti-CD20 antibody, has demonstrated activity against systemic PTLD.23, 24 However, the limited CNS penetration of rituximab has been a concern, because CSF concentrations reportedly were <5% of serum concentrations.25, 26 Nonetheless, case reports suggest that, despite limited concentration of the agent in the CNS, the drug has activity in this pharmacologic sanctuary.26, 27 In our series, patients who received rituximab-based regimens without concomitant cytotoxic chemotherapy or radiotherapy did very well, and 6 of 7 patients were alive at the 20-month follow-up. Efforts to improve CNS penetration, including dosage intensification and blood-brain barrier disruption, may enhance the activity of this agent.27, 28

Among patients with PTLD, it has been noted that CNS involvement is a poor prognostic factor.29, 30 Thus, it is surprising that the median survival was 47 months, exceeding the 26 months reported by Snanoudj et al.5 Furthermore, 34% of our patients remained alive at a median of 23 months follow-up. This more closely approximates outcomes among immunocompetent patients with CNS lymphoma. This finding suggests that patients with PCNS-PTLD should be considered for treatment, because outcomes are not uniformly poor.

Our study has several significant limitations related principally to the retrospective nature of the review. Because patients were identified through neuro-oncology programs, transplantations often were performed outside of the PTLD-treating institution, and detailed data were not available. Pathology was not reviewed centrally or reclassified according to modern classification systems. Similarly, imaging was not rereviewed. Treatments were not standardized and varied among the individual patients; furthermore, accurate palliation and toxicity assessments were not recorded routinely. Thus, these results must be interpreted with caution.

PCNS-PTLD is an uncommon complication of solid organ transplantation. Disease typically is multifocal and enhances in either a ring-enhancing or homogenous pattern. Diagnosis generally requires a cerebral biopsy, because supplementary assessments, including CSF and neuro-ophthalmogoic evaluations, usually are nondiagnostic. Tumors typically are B-cell, EBV-induced lymphomas. Several therapeutic options are available, although none have demonstrated clear superiority. Immune reconstitution, corticosteroids, and antiviral agents are used routinely, although their role remains unclear. Response rates to radiotherapy and chemotherapy are favorable, although the latter may be associated with less neurotoxicity. Alternatively, patients who received rituximab as primary treatment had surprisingly good outcomes. Regardless of the therapy administered, patient survival was superior to what has been reported previously. Given the rarity of this condition, collaboration is critical to improve our understanding of PCNS-PTLD and to better define optimal therapy.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.