In most cases of PTLD, ISR is the first step in treatment. ISR should partially restore the ability of CTLs to eliminate EBV-infected lymphocytes [8]. There are, however, several potential drawbacks associated with ISR—the most significant being graft rejection. In addition, ISR as monotherapy has a relatively slow time to response (on average>2–4 weeks) [65] and lower efficacy compared to ISR combined with rituximab and/or chemotherapy. Reported overall response rates (ORR) to ISR vary from 0 to 74 % [64, 66–68], with durable responses of less than 30 % of patients in several studies [13, 64, 66, 67].

In older studies, patients with early lesions and polyclonal PTLD appeared to respond better to ISR, while those with monoclonal tumors [46], in particular with bcl–6 expression, were less likely to respond [69, 70]. However, in a more recent study, monomorphic versus polymorphic histology did not predict for response to ISR [66].

In theory, EBV-positive PTLD should have a higher likelihood to respond to ISR; however, recent evidence suggests this may not necessarily be the case. In a recent study of SOT patients, EBV positivity was not a predictor of response—with only non-bulky disease (<7 cm) and age <50 at diagnosis being predictive [66]. Some factors such as high LDH and multi-organ dysfunction or involvement have inconsistently been associated with poor response to ISR [65, 66]. Multiple factors can be combined to help predict response to ISR. For example, using LDH elevation, hepatitis C infection, bone marrow or liver involvement, and B symptoms, 3-year overall survival (OS) was 100 % in patients with none of these factors, 79 % with one, and 8 % with two or more factors [66].

The specific dose modifications for ISR in an individual patient are based on multiple factors, such as the transplanted organ, extent of PTLD, perceived risk of rejection, symptoms, physical exam, and laboratory data. Decisions regarding ISR should be pursued in close collaboration with the transplant team. Anti-proliferative agents such as azathioprine and mycophenolate should be dose-reduced or discontinued, if possible [11]. Guidelines suggest dose reduction of 25 % in limited disease. With extensive disease in non-critically ill patients, cyclosporine/tacrolimus is typically reduced to 50 %, with discontinuation of azathioprine/mycophenolate and continuation of prednisone at 7.5–10 mg daily [71]. In critically ill patients, it may be necessary to discontinue all immune suppressing agents except prednisone 7.5–10 mg daily. Patients need to be followed closely to assess disease response and to monitor for graft rejection. Though there are risks with ISR, the majority of evidence supports the use of ISR as the first step in management of PTLD in most cases. In patients with aggressive disease, low predicted response to ISR alone or a contraindication to ISR may necessitate consideration of alternative or augmented therapies.

Rituximab, a chimeric monoclonal antibody targeting the B-cell surface protein CD20, has improved outcomes in many B-cell lymphomas. Data showing rituximab to be superior to other treatment approaches started to emerge around 2005 [72]. Prior to the rituximab era, the 3-year OS for PTLD ranged from 30 to 50 % but has now improved to >60 % [62, 72–75].

In the first trial of rituximab monotherapy use after ISR failure in PTLD, an ORR of 44 % was seen, with a median OS of 15 months [74]. Since then, several other studies have shown ORR ranging from 44 to 68 % and median OS up to 42 months (Table 2).

More recently, in a large, multicenter retrospective analysis, Evens et al. reviewed 80 SOT PTLD patients to establish the impact of the introduction of rituximab on outcomes [61]. All patients in the study had ISR, with 74 % of patients receiving rituximab with or without chemotherapy. With regard to first-line therapy, patients had a 73 % OS when rituximab was a part of the regimen compared to 33 % without rituximab. The 3-year progression-free survival (PFS) was 70 % with rituximab and 21 % without.

Prior to rituximab, chemotherapy was the primary treatment in patients who failed ISR. Several regimens have been described for use in PTLD. Anthracycline-based chemotherapy with rituximab, such as R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) is the most widely used regimen [71, 76–78]. It is preferred in patients for whom a rapid clinical response is required, such as those with multi-organ involvement or rapidly declining clinical status. Although the risk of toxicity and treatment-related mortality (TRM) can be as high as 25–50 % [79], the ORR is higher than rituximab monotherapy, ranging between 65 and 100 % [71]. Given the increased risk of TRM, chemotherapy should be reserved for those who fail to achieve appropriate response with ISR or ISR and rituximab, who have highly aggressive histology, or who require a rapid clinical response such as patients presenting with severe organ dysfunction or rapidly declining clinical status [71]. In some scenarios, it may be prudent to avoid certain chemotherapy drugs if possible, such as anthracyclines in heart transplant patients and high-dose methotrexate in renal transplant patients.

Many experts recommend a sequential or staged approach to the treatment of PTLD, in which patients generally receive ISR or ISR + rituximab as a starting point. An assessment of response is made 4–8 weeks later and, if the response is insufficient, additional therapy is then administered [63, 71, 80, 81]. This strategy was recently tested in a large prospective study (the “PTLD-1” trial) and may provide the foundation for a standard approach in the future [81]. In this study, 70 patients who failed ISR were then treated with 4 weekly doses of rituximab followed by 4 cycles of CHOP. Following rituximab, there was a 60 % ORR (with 20 % CR). After completion of rituximab × 4 and CHOP × 4, the ORR increased to 90 % (with 68 % CR) [81]. In this study, the planned therapy was rituximab × 4 followed by CHOP × 4, so it is unclear whether the patients who achieved CR with rituximab alone might have maintained their responses without chemotherapy.

In practice, a commonly utilized approach is to initially implement ISR or ISR + rituximab and then, for those who have not achieved CR, to subsequently apply chemotherapy. The “sequential therapy” approach of the PTLD-1 trial was modified to test this “risk-adapted” approach, after an interim analysis demonstrated that response to rituximab correlated with OS [82]. In the “risk-adapted” approach, patients achieving CR with 4 weeks of rituximab received 4 additional doses of rituximab, whereas those who did not achieve CR were then treated with 4 cycles of R-CHOP. In a preliminary analysis (n = 90), an ORR of 93 % and CR rate of 78 % were seen. The CR rate to rituximab alone was 27 %, with an additional 51 % going on to achieve CR after CHOP. Only 13 % of patients achieving CR with rituximab alone went on to relapse. Treatment-related mortality was low at 8 %. Comparing to previous data from the “sequential therapy” part of the PTLD-1 trial, it appears that (1) omitting chemotherapy from those achieving CR to rituximab is safe, and (2) for those who have progressive disease with rituximab, R-CHOP is more effective than CHOP. Those who failed to achieve CR after rituximab × 4 ± R-CHOP × 4 (22 %) did not benefit from additional R-CHOP and had nearly 80 % risk of progressive lymphoma in the following 1–2 years. Therefore, with this approach, most patients were able to achieve CR, while the relatively high treatment-related toxicity seen with CHOP was avoided except for patients who truly required chemotherapy [82]. Given these promising results, further study may help establish this “response-adapted” approach as a standard of care in CD20-positive PTLD [63].

Localized therapy using either radiation therapy (RT) or surgery can be utilized in carefully selected PTLD patients. When combined with ISR, this may be curative in patients with PTLD localized to a single site (Ann Arbor Stage I disease) [63]. RT may also have a significant role in the treatment of CNS or limited-stage PTLD, with some studies demonstrating complete responses [83, 84]. Incorporation of RT may also allow for an abbreviated course of chemotherapy for those with limited-stage disease or avoidance of chemotherapy for those who are unlikely to tolerate chemotherapy.

Primary CNS PTLD is unique in both its presentation and treatment compared to other forms of PTLD. CNS involvement occurs in up to 15–22 % of PTLD, most of which are primary CNS PTLD, and has consistently been associated with poor outcomes across multiple studies [6, 12, 61, 75, 84–87].

In a recent multicenter international retrospective analysis of 84 cases of primary CNS PTLD over a 14-year period, it was found that although 83 % of cases occurred late (>1 year post-transplant), over 90 % were EBV positive [87]. The large majority (79 %) were associated with renal transplantation, presented with a median time to onset of 54 months, and 79 % had diffuse large B-cell lymphoma (DLBCL) histology. Median PFS and OS for the cohort were 8 months and 17 months, respectively.

There are no clear guidelines for the treatment of primary CNS PTLD. Options include the use of ISR, rituximab, high-dose methotrexate (HD-MTX), high-dose cytarabine, and whole brain radiation [63]. The role of rituximab is poorly defined in this setting, given its poor penetration across the blood brain barrier [71]. Response to first-line therapy is the most important prognostic factor [85, 87]. In a review of 289 patients with primary CNS PTLD after SOT, it was reported that 32 of the 39 cases of CR occurred in patients who received RT [84]. Overall, however, the outcomes were poor, with <20 % of patients achieving long-term remission. In general, treatment with ISR alone is rarely effective for primary CNS PTLD. For now, treatment should be approached in a manner akin to the primary CNS lymphoma, including whole brain RT, HD-MTX, and high-dose cytarabine, bearing in mind that HD-MTX is likely to have increased toxicity in renal transplant patients [71, 87, 88].

Burkitt PTLD accounts for less than 5 % of PTLD cases [63]. It is a highly aggressive malignancy with frequent extra-nodal manifestations, high proliferative index, high latent EBV infection, and rapid tumor growth. Translocations involving c–myc and the heavy or light chain immunoglobulin loci underlie the aggressive biology. Burkitt Lymphoma, in the non-transplant population, has been successfully treated with short-duration intensive combination chemotherapy, along with aggressive CNS prophylaxis [89].

Unfortunately, there are no consensus guidelines or treatment protocols for the treatment of Burkitt PTLD. Due to the aggressive nature of the disease, ISR alone or ISR + rituximab are not recommended. Instead, most clinicians proceed directly to a combined approach involving ISR, rituximab, and chemotherapy. In a small study, CR was achieved in 5/5 patients prospectively treated with ISR and R-CHOP [90]. The role of intrathecal chemoprophylaxis is unclear in Burkitt PTLD. However, extrapolating from the non-transplant setting, we recommend CNS prophylaxis in cases of Burkitt PTLD as well. Highly aggressive regimens typically used for BL in the non-transplant setting should be used with extreme caution in PTLD patients due to the increased risk for toxicity [91].

Hodgkin PTLD is the rarest form of PTLD; as a result, treatment data are limited to a few case reports and series [92, 93]. The approach to treatment for Hodgkin PTLD is similar to that employed in Hodgkin’s lymphoma (HL). This typically involves chemotherapy or combined modality treatment using chemotherapy and RT [94]. The most commonly used chemotherapy regimen for HL in the USA is ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine). In one study of 7 patients with Hodgkin PTLD treated with standard HL therapy along with ISR, 4 patients achieved CR [93]. In another series, 4 out of 4 patients treated with standard regimens achieved CR [92].

Anti-viral agents, such as acyclovir and ganciclovir, have been studied as either prophylaxis or treatment for PTLD. These agents require activation by EBV thymidine kinase (EBV-TK), an enzyme expressed in infected replicating cells, and are therefore ineffective against latent EBV-infected B cells [95, 96]. Arginine butyrate can be used to induce EBV-TK in infected latent B cells, with subsequent administration of ganciclovir to eliminate these cells. Results utilizing arginine butyrate and ganciclovir are encouraging [95], but further trials are required to prove clear clinical benefit. Prophylactic use of these agents has had mixed results [71]. Outside of clinical trials, anti-viral agents are not currently recommended for routine prophylaxis or treatment of PTLD.