Successful Treatment of Primary Central Nervous System Post-Transplant Lymphoproliferative Disorder (PCNS-PTLD) with Zidovudine (AZT), Ganciclovir (GCV), Rituximab and Dexamethasone: A Single-Institution Case Series

Primary central nervous system post-transplant lymphoproliferative disorder (PCNS PTLD) is a rare complication of solid organ transplantation, with no standard therapy. Most PCNS PTLDs are associated with Epstein-Barr virus (EBV) infection, thus EBV could serve as a potentially attractive therapeutic target. For EBV-targeted antiviral therapy to be effective, antiviral agents must be phosphorylated by lytic-phase EBV kinases BXLF1/vTK and BGLF4. We hypothesized that PCNS-PTLDs would express viral kinases and that antiviral therapy would prove an attractive therapeutic alternative to high dose methotrexate-based chemotherapy for patients who often have impaired organ function and poor performance status.

We investigated the safety and efficacy of EBV-targeted therapy with zidovudine (AZT), ganciclovir (GCV), rituximab, and dexamethasone in patients with PCNS PTLD, as well as the relationship of viral protein kinase expression with response to therapy. Patients with biopsy-proven PCNS PTLD following solid organ transplantation were eligible for treatment. Induction therapy consisted of AZT 1500 mg IV, and GCV 5 mg/kg IV, dexamethasone 10 mg IV, twice daily on days 1–14 and 4 weekly doses of rituximab 375 mg/m2 on days 1, 8, 15, and 22. Maintenance therapy was initiated on day 15 with valganciclovir 500 mg twice daily, and AZT 300 mg twice daily until disease progression or intolerable toxicity. Treatment was adjusted for hematologic toxicity and impaired liver and kidney function. Responses were evaluated by serial brain MRIs beginning 4 weeks after initiation of treatment. Brain biopsy specimens were evaluated for expression of viral kinases BGLF4 and BXLF1/vTK by in situ hybridization.

Eight patients (5 M, 3 F) with a median age of 49 were treated at our institution from 1999–2011. Transplant history included kidney (N=7), and kidney + pancreas (N=1). Pathology data was available for all patients and included diffuse-large B-cell lymphoma (N=4), grade III lymphomatoid granulomatosis (N=2), and B-cell lymphoma, not further classifiable (N=2). EBV positivity (EBER) and CD20 expression was documented in all cases. Evaluation of BXLF1/vTK and BGLF4 was completed in 4 patients and found to be positive. Areas of tumor expressing viral kinases did not express LMP1. Immune suppression was reduced in all patients prior to treatment. All patients completed induction therapy. Median duration of maintenance therapy was 16.6 months. At the time of analysis, 6 patients were still alive and disease free (median duration of follow-up = 19.6 months, inter-quartile range = 10.6 – 28.3). All 8 patients achieved a complete response by MRI criteria with a median duration to response of 2 months (inter-quartile range = 1.5 – 4 months). Two patients died with survival times of 3 and 143 months after diagnosis. No patients had documented disease progression. Thus far, median duration of progression free survival has been 17.1 months (inter-quartile range = 7 – 31 months). Patient 1 was disease-free for 141 months but developed and died from complications of colon cancer. Patient 8 died of multi-organ failure related to pneumonia and septic shock, which was considered non-treatment-related, 3 months after initiation of therapy. At that time, a brain MRI showed evidence of complete response. Grade 3–4 toxicity was primarily hematologic, including anemia (N=4), thrombocytopenia (N=3), leukopenia (N=5) and neutropenia (N=4). Toxicity in the maintenance phase was generally reversible within 7 days of holding therapy. One patient required discontinuation of AZT during maintenance treatment for persistent, transfusion-dependent anemia.

EBV-targeted therapy with AZT, GCV, rituximab and dexamethasone appears to be safe for the treatment of EBV+ PCNS PTLD, with promising evidence of activity. Responses were generally rapid and durable. Expression of BXLF1/vTK and BGLF4 kinases provides mechanistic rationale for an antiviral approach for this disease. Given the lack of effective standard therapy for these patients, this regimen deserves further investigation. Unanswered questions include the optimum length of maintenance therapy that is still effective while minimizing toxicity, and whether rituximab is necessary to achieve the responses observed. A multi-center phase II trial is in development to further investigate this regimen.

Off Label Use: We will discuss the use of zidovudine and ganciclovir to treat primary CNS lymphoma in transplant patients.