Journal Logo

Original Clinical Science—General

Real-world Outcomes With Rituximab-based Therapy for Posttransplant Lymphoproliferative Disease Arising After Solid Organ Transplant

Burns, David M. PhD1,2; Clesham, Katherine MB ChB3; Hodgson, Yan A. MBBS4; Fredrick, Lynsey MB ChB5; Haughton, Joanna MB ChB6; Lannon, Michelle MB ChB7; Hussein, Hayder BMBS7; Shin, Jin-Sup MBBS3; Hollows, Robert J. PhD2; Robinson, Lisa MBBS8; Byrne, Catherine MBBS9; McNamara, Christopher MBBS3; Vydianath, Bindu MBBS10; Lennard, Anne L. MBBS7; Fields, Paul PhD11; Johnson, Rod MD6; Wright, Josh MD5; Fox, Christopher P. PhD4; Cwynarski, Kate PhD3; Chaganti, Sridhar PhD1

Author Information
doi: 10.1097/TP.0000000000003183

Abstract

INTRODUCTION

Posttransplant lymphoproliferative disease (PTLD) remains a serious complication of solid organ transplantation, occurring in 5%–10% of patients and conferring a high risk of mortality.1,2 The disease arises due to loss of immune surveillance consequent upon iatrogenic immunosuppression and is commonly associated with Epstein-Barr virus (EBV), a B-lymphotropic herpes virus which drives tumorigenesis in the absence of effective virus-specific T-cell responses.3 PTLD displays marked pathological heterogeneity, although the majority of cases are CD20+ monomorphic B-cell lymphomas resembling diffuse large B-cell lymphoma or polymorphic lesions characterized by a spectrum of B-lymphoid cell types.4

In the first-line management of PTLD, reduction of iatrogenic immunosuppression (RI) is often undertaken with the aim of stimulating endogenous antitumoral immune responses, although this is frequently inadequate to produce sustained remission in established PTLD and exposes patients to the risk of organ rejection. Consequently, upfront therapy for B-cell PTLD typically employs the anti-CD20 antibody Rituximab, either as monotherapy or in combination with cytotoxic chemotherapy, typically with CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisolone). Rituximab monotherapy (R-Mono) has been evaluated in a small number of prospective series,5-8 delivering overall response rates of 44%–79% in unselected patients, although a lack of durable remissions has been recognized as a limiting factor. Meanwhile, chemotherapy has been associated with improved response rates and attainment of durable remissions.9-12 However, some authors have raised concerns about the toxicity of CHOP chemotherapy in this patient group, with a treatment-related death rate of 31% reported in 1 series.13 In view of this, a strategy involving upfront R-Mono for all patients, followed by escalation to chemotherapy in those with suboptimal response, was evaluated in the phase II PTLD-1 study,14,15 demonstrating an overall response rate of 88% and median overall survival (OS) of 6.6 years. This approach is regarded by some as a new standard of care in the treatment of PTLD, although a direct comparison between upfront R-Mono and Rituximab plus CHOP (R-CHOP) is lacking.

Further complicating the management of PTLD has been the lack of a well-validated disease-specific prognostic index to guide therapeutic decision-making. Several prognostic scoring systems have been proposed, although these have been derived from heterogeneous cohorts which have included a spectrum of pathologies and differing types of therapy, limiting their reproducibility and clinical applicability.16-19 The International Prognostic Index (IPI) for high-grade non-Hodgkin lymphoma 20 has been shown to be prognostic in several studies of PTLD.17-19,21,22 However, disease characteristics, such as a high proportion of cases with advanced stage and extranodal involvement at presentation, have the potential to limit the prognostic utility of the IPI for PTLD,16,18 and therefore additional validation is required in this setting.

Considering these issues, we undertook a multicenter study to capture real-world practice in the management of B-cell PTLD. Performing a comparative analysis of R-Mono and R-CHOP, we sought to explore factors that influence choice of therapy and report corresponding response and survival outcomes. Furthermore, we wished to assess the applicability of the IPI to a cohort of patients treated exclusively with upfront R-Mono or R-CHOP.

MATERIALS AND METHODS

This multicenter study invited participating centers to submit data on adult patients (aged 16 y and over) with biopsy-proven PTLD arising after solid organ transplant. Data were collected from 7 transplant centers in the United Kingdom. All clinical, radiological, and pathological data were critically reviewed and the following inclusion criteria were applied: biopsy-proven PTLD; preceding solid organ transplant; treated with R-Mono or R-CHOP as upfront therapy; diagnosed between 2000 and 2015. The following exclusion criteria were applied: not biopsy-proven; subtypes of PTLD other than polymorphic, diffuse large B-cell lymphoma or B-cell PTLD not otherwise specified; treatment with another form of chemotherapy before R-Mono or R-CHOP; failure to provide a minimum dataset necessary for analysis of survival outcomes. The study was conducted in line with National Research Ethics Service guidance and was registered with Research and Development departments at each center.

Pathology reports were used to categorize cases according to World Health Organization criteria for PTLD.4 Cases were considered to be EBV-associated if Epstein-Barr virus-encoded small RNAs in situ hybridization and/or latent membrane protein-1 immunohistochemistry were positive; negative staining for latent membrane protein-1 in the absence of Epstein-Barr virus-encoded small RNAs in situ hybridization was not considered sufficient to exclude EBV-association and such cases were classified as equivocal. Early-onset PTLD was defined as that occurring within 12 months of organ transplantation. Disease stage was determined from available clinical, bone marrow and radiological data, the latter comprising computed tomography (CT) and/or positron emission tomography (PET),23-25 in accordance with the Ann Arbor staging system,26,27 including the Lugano classification for primary gastrointestinal lesions.28

Patients were treated as per physician preference, with supportive therapy delivered according to local protocols. Response was evaluated on the basis of submitted clinical and radiological data and was categorised into complete response (CR), partial response (PR) or no response (NR), in line with the Cheson criteria where response was assessed by CT or PET.29 Owing to the nature of the study, the timing and modality of response evaluation were determined clinically. OS was calculated from the date of diagnosis to the date of death or last follow-up. Progression-free survival (PFS) was calculated from date of diagnosis to date of disease progression or death from any cause. Association between baseline variables and treatment type was analyzed using binary logistic regression. Other categorical variables were compared with chi-square test or Fisher exact test as appropriate. Kaplan-Meier curves were compared with Log Rank testing. Prognostic factors for OS were analyzed using Cox proportional hazards modeling. All tests of statistical significance used 2-tailed testing and assumed a significance level of 0.05. Statistics were performed using SPSS (IBM) and Stata (StataCorp LLC).

RESULTS

Patient Characteristics

One-hundred one patients with biopsy-proven B-cell PTLD treated with upfront R-Mono or R-CHOP were included in this analysis from a total of 163 cases initially identified (Table 1; Figure S1, SDC, https://links.lww.com/TP/B884). The median age at diagnosis was 49 years (range, 16–84 y) and a male predominance (69%) was observed. Most patients had received either a renal (61%) or liver (32%) transplant. The median time from transplant to PTLD diagnosis was 89 months (range, 2–302 mo), with early-onset disease noted in 18% of cases. All patients underwent staging with CT, and/or PET in 49 of 101 (49%) cases, with stage III–IV disease documented in 52 of 101 (52%) patients at presentation. Extranodal disease is common in PTLD and consistent with this 75 of 101 (74%) patients exhibited involvement of at least 1 extranodal site, with 26 of 101 (26%) having 2 or more involved sites. The gastrointestinal tract was the most commonly involved extranodal site, observed in 44 of 75 (59%) of those with extranodal disease. EBV-association was demonstrated in 50% of satisfactorily evaluated cases and, as anticipated, this was more common in early-onset (P = 0.00004) and polymorphic (P = 0.001) disease.

TABLE 1. - Characteristics of patients treated with R-Mono or R-CHOP
Variable R-Mono R-CHOP Total ORa P
Patients, no. 41 (40.6%) 60 (59.4%) 101 (100.0%)
Age, y
 Median 50 49 49 1.00 0.828
 Range 16–84 17–72 16–84
Sex
 Male 30/41 (73.2%) 40/60 (66.7%) 70/101 (69.3%) 1.36 0.487
Transplant type
 Renal 25/41 (61.0%) 37/60 (61.7%) 62/101 (61.4%) 0.976
 Liver 14/41 (34.1%) 18/60 (30.0%) 32/101 (31.7%)
 Cardiothoracic 1/41 (2.4%) 4/60 (6.7%) 5/101 (5.0%)
 Other 1/41 (2.4%) 1/60 (1.7%) 2/101 (2.0%)
Onset
 Early 15/41 (36.6%) 3/60 (5.0%) 18/101 (17.8%) 10.96 <0.001
 Late 26/41 (63.4%) 57/60 (95.0%) 83/101 (82.2%)
PTLD subtype
 Polymorphic 13/41 (31.7%) 4/60 (6.7%) 17/101 (16.8%) 7.18 0.002
 DLBCL 24/41 (58.5%) 53/60 (88.3%) 77/101 (76.2%)
 B-cell not otherwise specified 4/41 (9.8%) 3/60 (5%) 7/101 (6.9%)
EBV
 Positive 27/36 (75.0%) 12/42 (28.6%) 39/78 (50.0%) 7.50 <0.001
Stage
 I–II 21/41 (51.2%) 28/60 (46.7%) 49/101 (48.5%)
 III–IV 20/41 (48.8%) 32/60 (53.3%) 52/101 (51.5%) 0.83 0.653
Extranodal disease
 ≥1 site 31/41 (75.6%) 44/60 (73.3%) 75/101 (74.3%) 1.13 0.797
 ≥2 sites 5/41 (12.2%) 21/60 (35.0%) 26/101 (25.7%) 0.26 0.014
B symptoms
 Present 15/34 (44.1%) 27/50 (54.0%) 42/84 (50.0%) 0.67 0.375
LDH
 Elevated 23/39 (59.0%) 31/54 (57.4%) 54/93 (58.1%) 1.07 0.880
ECOG PS
 0–2 33/40 (82.5%) 53/57 (93.0%) 86/97 (88.7%)
 3–4 7/40 (17.5%) 4/57 (7.0%) 11/97 (11.3%) 2.81 0.120
IPI
 Low (0–2) 23/38 (60.5%) 31/53 (58.5%) 54/91 (59.3%) 1.09 0.845
 High (3–5) 15/38 (39.5%) 22/53 (41.5%) 37/91 (40.7%)
aOdds ratio for selection of R-Mono vs R-CHOP as initial therapy for PTLD. Numbers in parentheses indicate percentages.
DLBCL, diffuse large B-cell lymphoma; EBV, Epstein-Barr virus; ECOG PS, Eastern Cooperative Oncology Group Performance Status; IPI, International Prognostic Index; LDH, lactate dehydrogenase; OR, odds ratio; PTLD, posttransplant lymphoproliferative disease; R-CHOP, Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone; R-Mono, Rituximab monotherapy; –, odds ratio not determined.

Upfront Treatment

Initial therapy was with R-Mono in 41 of 101 (41%) and R-CHOP in 60 of 101 (59%). Detailed information regarding RI was not obtained, but R-Mono or R-CHOP were introduced concurrently with RI in the majority of cases. Of patients treated with R-Mono, 40 of 41 (98%) completed at least 4 weekly infusions of Rituximab 375 mg/m2 and 9 of 41 (22%) received 8 infusions. Of those treated upfront with R-CHOP, 45 of 60 (75%) completed at least 6 cycles.

Factors Influencing Choice of Upfront Therapy

Statistical analysis was performed to understand the baseline factors that may have influenced choice of initial therapy (Table 1). This revealed that R-Mono was used more frequently in patients with polymorphic histology (odds ratio [OR], 7.18; P = 0.002), with early-onset disease (OR, 10.96; P < 0.001) and with EBV-positive disease (OR, 7.50; P < 0.001). Meanwhile, it was used less frequently in those with ≥2 extranodal sites (OR, 0.26; P = 0.014).

Outcomes

Response was assessed by PET (52/101 [52%]) or CT (40/101 [40%]) in all cases, except for 3 patients with gut involvement who were assessed by repeat endoscopic evaluation, 3 patients with disease progression determined clinically, and 3 patients with early deaths on treatment. Among all patients, compared with R-Mono, there was a trend towards improved overall (75.0% versus 89.7%; P = 0.054) and complete (52.5% versus 70.7%; P = 0.066; Table 2) response rates with R-CHOP. Response assessment by PET was associated with a higher rate of CR than with CT, with 12 of 21 (57%) and 7 of 17 (41%) of patients treated with R-Mono, and 26 of 31 (84%) and 13 of 23 (57%) patients treated with R-CHOP, achieving CR, respectively.

TABLE 2. - Response to R-Mono or R-CHOP
Response R-Mono (N = 40a) R-CHOP (N = 58a) P
OR 30 (75.0%) 52 (89.7%) 0.054
CR 21 (52.5%) 41 (70.7%) 0.066
PR 9 (23%) 11 (19.0%)
NR 10 (25.0%) 6 (10.3%)
aOne case treated with R-Mono and 2 cases treated with R-CHOP were not evaluable for response due to early deaths.
CR, complete response; NR, no response; OR, overall response; PR, partial response; R-CHOP, R-Mono or Rituximab plus CHOP, doxorubicin, vincristine, and prednisolone; R-Mono, Rituximab monotherapy.

Among those treated with R-Mono, 25 of 41 (61%) had no further therapy and remained relapse or progression free with a median follow-up of 25 months. Eleven other patients who received R-Mono were switched to R-CHOP after completion of R-Mono for refractory disease (6/11), for response less than CR (4/11), or for consolidation of CR (1/11). Two other patients were switched to Rituximab plus CHOP (R-CHOP), vincristine, and prednisolone or Rituximab plus CHOP (R-CHOP) and prednisolone for refractory disease.

With a median follow-up of 47 months, the 1-year and 3-year OS estimates for the whole cohort were 79.8% (95% confidence interval [CI], 70.4%-86.5%) and 65.9% (95% CI, 55.1%-74.6%) respectively, with a median OS of 87 months (Figure 1A). One-year and 3-year PFS for the whole cohort was 73.0% (95% CI, 62.9%-80.8%) and 67.3% (95% CI, 56.8%-75.8%) respectively, with median PFS of 87 months (Figure 1B). Notably, PFS and OS rates were indistinguishable by 3 years, reflecting the poor outlook for patients with disease progression or relapse. There was no significant difference in OS for patients treated upfront with R-Mono versus R-CHOP, with 1-year OS of 82.7% (95% CI, 67.1%-91.4%) and 77.8% (95% CI, 64.8%-86.5%), and 3-year OS of 70.6% (95% CI, 52.4%-82.9%) and 62.5% (95% CI, 48.4%-73.8%), respectively (P = 0.722; Figure 2A). Similarly, there was no significant difference in PFS between R-Mono and R-CHOP, with 1-year PFS of 80.1% (95% CI, 64.1%-89.5%) and 68.1% (95% CI, 54.2%-78.6%) and 3-year PFS of 72.1% (95% CI, 55.2%-83.5%) and 64.2% (95% CI, 50.1%-75.3%), respectively (P = 0.627; Figure 2B).

F1
FIGURE 1.:
Survival among all patients with posttransplant lymphoproliferative disease (PTLD). A, Overall and (B) progression-free survival among patients with PTLD arising after solid organ transplant treated upfront with either Rituximab monotherapy (R-Mono) or Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP) chemotherapy.
F2
FIGURE 2.:
Survival by Rituximab monotherapy (R-Mono) or Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP) upfront therapy. A, Overall and (B) progression-free survival among patients with posttransplant lymphoproliferative disease (PTLD) arising after solid organ transplant treated upfront with R-Mono or R-CHOP chemotherapy.

Toxicity

Among those treated with R-Mono, 10 of 41 (24%) patients died within 12 months of initiating treatment. Three patients died from non-PTLD causes, in remission after only having received R-Mono: a patient who died from necrotizing fasciitis with respiratory failure; a patient with a hepatic abscess and hepatic artery thrombosis; and a patient with liver rejection. Two other patients died in the context of disease progression without further therapy. Five other patients died after switching to a second line of therapy, 4 of whom died in the context of disease progression, while the remaining patient died from sepsis with unknown disease status having received 6 cycles of R-CHOP after completing R-Mono. In summary, non-PTLD mortality within 12 months of completing upfront R-Mono was apparent in 3 of 41 (7%) patients.

Among those treated with upfront R-CHOP, 17 of 60 (28%) patients died within 12 months of initiating treatment. Of these, 5 of 60 (8%) died “on treatment” (defined as death within 30 d of the last infusion), 2 of whom who had achieved remission: a patient who died from subarachnoid hemorrhage 3 weeks after completing 6 cycles of R-CHOP, having attained a CR on imaging after 4 cycles; and a patient who died 1 week after completing 6 cycles of R-CHOP from neutropenic infection, having attained PR on imaging performed after 4 cycles. Two other patients who died on treatment had clinically evident disease progression: a patient who died after 1 cycle of R-CHOP with neutropenic infection, and a patient who died after 4 cycles with infection. One other patient died on treatment from portal vein thrombosis and venous bowel infarction after 1 cycle of R-CHOP, presumably as a complication of PTLD. Of the 12 other patients who died within 12 months, 1 patient failed to tolerate R-CHOP and was deescalated to Rituximab plus CHOP (R-CHOP), vincristine, and prednisolone and died with apparently stable disease from a non-PTLD cause, and 1 patient achieved CR with R-CHOP but subsequently died from an unknown cause. Ten other patients died in the context of overt disease progression. In summary, non-PTLD mortality within 12 months of completing frontline R-CHOP was apparent in 4 of 60 (7%) patients.

Prognostic Analysis

Prognostic factors for OS were assessed using Cox proportional hazards analysis (Table 3). Among patients treated with either R-Mono or R-CHOP, significant baseline predictors of inferior OS included age >60 years (hazard ratio [HR], 2.88; 95% CI, 1.49-5.56; P = 0.002), stage ≥3 disease (HR, 2.02; 95% CI, 1.04-3.92; P = 0.04), ≥2 extranodal sites (HR, 2.37; 95% CI, 1.23-4.58; P = 0.01), elevated lactate dehydrogenase (HR, 2.47; 95% CI, 1.12-5.46; P = 0.026), and Eastern Cooperative Oncology Group Performance Status ≥3 (HR, 2.89; 95% CI, 1.25-6.68; P = 0.013). However, histology, EBV-association, the presence of extranodal disease, time of onset from transplant, and patient sex were not significantly prognostic. Notably, both overall response (HR, 0.21; 95% CI, 0.10-0.44; P = 0.00004) and complete response (HR, 0.26; 95% CI, 0.13-0.52; P = 0.0001) to initial therapy were highly prognostic for OS.

TABLE 3. - Factors predicting overall survival with R-Mono or R-CHOP
Univariate analysis HR (95% CI) P
Age, y 1.04 (1.01-1.06) 0.001
Age >60 y 2.88 (1.49-5.56) 0.002
Male sex 1.16 (0.57-2.34) 0.684
Late onset 0.80 (0.35-1.83) 0.601
Polymorphic 0.92 (0.40-2.12) 0.852
EBV association 1.04 (0.51-2.15) 0.908
Stage ≥3 2.02 (1.04-3.92) 0.038
Extranodal disease 2.07 (0.86-4.95) 0.103
Extranodal sites ≥2 2.37 (1.23-4.58) 0.010
Elevated LDH 2.47 (1.12-5.46) 0.026
ECOG PS ≥3 2.89 (1.25-6.68) 0.013
CI, confidence interval; EBV, Epstein-Barr virus; ECOG PS, Eastern Cooperative Oncology Group Performance Status; HR, hazard ratio; LDH, lactate dehydrogenase; R-CHOP, Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone; R-Mono, Rituximab monotherapy.

Sufficient data were available to apply the IPI for high-grade non-Hodgkin lymphoma to 91 of 101 (90%) patients treated with R-Mono or R-CHOP. Patients were stratified according to individual risk scores with a high degree of statistical significance (P < 0.0001), although there was more limited separation of patients scoring 1, 2, or 3 points (Figure 3A). Furthermore, patients were separated into low- and high-risk groups according to whether they scored 0–2 points (N = 54) or ≥3 points (N = 37), consistent with prior application of the IPI in PTLD.22 The corresponding OS at 3 years for these groups was 77.5% (95% CI, 62.7%-87.0%) and 53.6% (95% CI, 35.6%-68.7%), respectively (P = 0.0003; Figure 3B).

F3
FIGURE 3.:
Survival by International Prognostic Index (IPI) group. Overall survival among patients treated upfront with Rituximab monotherapy (R-Mono) or Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP) was determined according to the IPI for high-grade non-Hodgkin lymphoma (NHL), with 1 point assigned for each of the following baseline factors: age >60 y, stage ≥3 disease, Eastern Cooperative Oncology Group Performance Status (ECOG PS) ≥2, extranodal sites ≥2, and elevated lactate dehydrogenase (LDH). A, Survival according to IPI score. B, Survival among low- (0–2 points) vs high-IPI (≥3 points) groups.

Applying the IPI, we subsequently explored the hypothesis that R-Mono may deliver similar outcomes to R-CHOP as upfront therapy for patients with low-risk PTLD (Table 4 and Figure 4). This analysis revealed similar overall (19/23 [82.6%] versus 27/31 [87.1%]; P = 0.711) and complete response (16/23 [69.6%] versus 22/31 [71.0%]; P = 0.911) rates for patients with low-risk PTLD treated with R-Mono and R-CHOP, respectively. In contrast, among patients with high-risk PTLD, R-Mono was associated with a borderline inferior overall response rate (9/14 [64.3%] versus 20/22 [90.9%]; P = 0.084) and a significantly inferior complete response rate (3/14 [21.4%] versus 15/22 [68.2%]; P = 0.006) compared with R-CHOP. A survival difference was not identified between therapies for patients with low- or high-IPI PTLD (Figure 4).

TABLE 4. - Outcomes with R-Mono or R-CHOP according to IPI risk group
Outcome Low-risk PTLD High-risk PTLD
R-Mono R-CHOP P R-Mono R-CHOP P
OR 19/23 (82.6%) 27/31 (87.1%) 0.711 9/14 (64.3%) 20/22 (90.9%) 0.084
CR 16/23 (69.6%) 22/31 (71.0%) 0.911 3/14 (21.4%) 15/22 (68.2%) 0.006
3-y OS 75.1% (48.6%-89.3%) 78.9% (58.7%-90.0%) 0.983 69.8% (37.8%-87.6%) 48.1% (26.1%-67.1%) 0.562
CR, complete response; IPI, International Prognostic Index; OR, overall response; OS, overall survival; PTLD, posttransplant lymphoproliferative disease; R-CHOP, Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone; R-Mono, Rituximab monotherapy.

F4
FIGURE 4.:
Overall survival by International Prognostic Index (IPI) group and upfront therapy. Overall survival among patients with (A) low- and (B) high-IPI posttransplant lymphoproliferative disease (PTLD) treated with upfront Rituximab monotherapy (R-Mono) or Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP).

DISCUSSION

In earlier studies, R-Mono was well tolerated as upfront therapy for PTLD, but durable remissions were achieved in less than half of unselected patients.5-8 Anthracycline-based chemotherapy, principally in the form of CHOP, with or without Rituximab, was associated with improved response and survival outcomes,9-12 although concerns were raised about its toxicity in this patient group.13,14 Consequently, Trappe et al14 introduced “sequential therapy,” with a protocol comprising frontline treatment with 4 weekly infusions of Rituximab, with the aim of improving performance status and tolerability, before delivering 4 subsequent cycles of CHOP-21. This strategy was evaluated in 70 patients in the phase II PTLD-1 study, which demonstrated an overall response rate of 60% following R-Mono, rising to 90% after CHOP, and a median OS of 6.6 years. The PTLD-1 trial protocol was thereafter amended to incorporate response-stratified treatment, such that patients with less than CR after 4 Rituximab infusions received 4 cycles of R-CHOP-21, but those who achieved CR were instead consolidated with 4 additional 3 weekly infusions of Rituximab.15 In the second part of the trial, 174 patients were treated with similar outcomes, with an overall response rate of 88% and median survival of 6.6 years, and 25% of the cohort were treated exclusively with R-Mono. The PTLD-1 trial is a major accomplishment, both as the largest prospective study delivered in PTLD, and with the demonstration that a response-stratified approach can facilitate chemotherapy-free cure in a quarter of patients. However, it is important to recognize that in the absence of a randomized comparison of upfront R-Mono and R-CHOP, it remains unknown to what extent upfront R-Mono genuinely offsets toxicity associated with R-CHOP, and indeed whether upfront R-Mono is the most beneficial initial therapy for all patients.

In the present study, we report outcomes among patients treated exclusively with either upfront R-Mono or R-CHOP, using real-world data in which therapy was selected at the discretion of treating physicians. The majority of patients had undergone renal or liver transplantation, with few recipients of heart or lung transplants, noting that thoracic organ transplantation has been identified as an independent risk factor for inferior outcome.15 With a 3-year OS of 65.9% and a median OS of 87 months for all patients, our data compare favorably with previous studies, including the PTLD-1 trial where 3-year OS was estimated as 61% (95% CI, 49%-72%) in the first phase,14 and 70% (95% CI, 62%-77%) in the second phase,15 although 25% of patients had received cardiothoracic transplants in PTLD-1. Meanwhile, in the retrospective study by Evens et al30 3-year OS of 73% (95% CI, 58%-83%) was reported among 59 patients treated with Rituximab-based therapy. Importantly, we also showed that 61% of patients treated upfront with R-Mono achieved long-term remission without any further therapy.

Comparing outcomes with upfront R-Mono or R-CHOP, we found inferior response rates with R-Mono, albeit with borderline statistical significance. However, this did not translate into a significant survival difference between therapies, presumably because patients who experienced a suboptimal response to upfront R-Mono were able to subsequently receive and benefit from R-CHOP—as such, these patients received a response-adapted escalation to chemotherapy similar to that applied in the PTLD-1 trial. Given the nonrandom allocation of therapy, we undertook statistical analysis of baseline variables to detect any potentially important differences between the groups. This revealed a preference for R-Mono in patients with polymorphic histology, although it is notable that monomorphic histological classification has not shown adverse prognostic impact in several prior studies,14-16,18,19,30 including the PTLD-1 trial and the present study. R-Mono was also preferred in cases of EBV-associated and early-onset disease, although these factors also lack prognostic significance in most prior studies; they are also not independent of histological type because polymorphic PTLD is typically EBV-associated and early-onset. R-Mono was also favored in those with <2 extranodal disease sites, conferring a potential survival advantage. However, the proportion of patients with high-IPI disease was not found to differ between upfront therapies. Notably, we also found an increased rate of CR among patients assessed by PET imaging. This may explain, at least partly, the relatively high proportion of patients who achieved CR with R-Mono in the present study, whereas in preceding trials of R-Mono response assessment was undertaken using CT imaging.5-8

Concerns have been raised about CHOP-associated toxicity in patients with PTLD. Thus, in the retrospective study by Choquet et al13 treatment-related mortality (TRM) was reported as 31% among 26 patients treated with CHOP-21. Meanwhile, in a retrospective analysis by Elstrom et al10 TRM was 26% among 19 patients treated with CHOP or R-CHOP. In the PTLD-1 trial, CHOP-related mortality was markedly lower at 11% and 8% in the first and second phases of the study respectively, and this improvement was largely attributed to better tolerability of CHOP resulting from prior Rituximab. Our own experience differs, however, because we did not observe an excess of CHOP-associated mortality. Thus, non-PTLD (7% versus 7%) and all-cause (24% versus 28%) mortality within 12 months of therapy were similar for those treated upfront with R-Mono and R-CHOP. Indeed, the majority of patients who died after R-CHOP did so from disease progression, rather than non-PTLD causes. This might be explained, perhaps, by careful selection of patients for R-CHOP or by improved supportive care. However, it is particularly notable that in the Choquet et al and Elstrom et al10,13 studies, a high proportion of patients (50% and 54%, respectively) had undergone cardiac or lung transplantation, making these higher-risk populations that were probably less able to tolerate R-CHOP. Therefore, we contend that CHOP-associated mortality may have been previously overestimated, and that with appropriate case selection and supportive care, excessive TRM may be avoidable.

A number of PTLD-specific prognostic scoring systems have been proposed, principally derived from analysis of small heterogeneous cohorts, although none have been widely adopted.16-19 Their utility was recently evaluated in a comparative analysis of patients treated prospectively in the PTLD-1 trial.22 This confirmed the prognostic significance of the Ghobrial17 and PTLD Prognostic Index18 scoring systems, although the authors concluded that none was superior to the IPI when considering the latter’s practical utility as a well-recognized and easily-applied tool. The IPI has been prognostic in at least 4 other studies of PTLD.17-19,21 In the present analysis, we confirmed the prognostic significance of the IPI among a cohort of patients treated exclusively with R-Mono or R-CHOP. Thus, in univariate analysis, all elements of the IPI (age >60 y, stage ≥3, Eastern Cooperative Oncology Group Performance Status ≥2, extranodal sites ≥2, and elevated lactate dehydrogenase) were significantly predictive for OS, and it performed with a high degree of statistical significance (P < 0.0001) in stratifying patients according to individual IPI scores, albeit with more limited discrimination of patients scoring 1, 2, or 3 points. Applying the IPI with a cutoff of ≥3 points to define high-risk patients, we again found the score to be discriminatory with a high degree of significance. Importantly, we used the IPI to compare outcomes with upfront R-Mono and R-CHOP in low- and high-risk groups. This revealed that in patients with low-IPI disease, R-Mono and R-CHOP were equivalent, delivering similar response and survival outcomes. In contrast, among those with high-IPI disease, R-Mono conferred inferior complete response rates compared with R-CHOP, although this did not translate into a survival difference. This analysis confirms that in a subset of patients with high-risk disease, R-Mono is less effective in delivering response. However, this does not appear to impact survival because patients with suboptimal responses after R-Mono can subsequently receive R-CHOP. Accordingly, our data support the use of R-Mono as the preferred upfront therapy for the majority of patients with PTLD, on the basis that a proportion will achieve cure exclusively with Rituximab, accepting that the remainder will require subsequent escalation to R-CHOP. Nonetheless, in a minority of high-risk cases, such as those with widespread bulky disease or who are clinically unstable at presentation, it might be unsafe to risk an inadequate response with R-Mono and delay R-CHOP. For such patients, where there is a need to deliver response rapidly, we contend that upfront R-CHOP remains a valid therapeutic option, particularly as we did not observe an excess of CHOP-associated mortality in this study.

Our study has several limitations, and therefore caution should be exercised in interpreting the findings. Thus, as a retrospective analysis, our data are liable to multiple biases. Even with 101 patients, the study is limited by small sample size. The cohort was predominantly composed of renal and liver transplant recipients, with few cardiothoracic transplant patients. Data presented on toxicity is limited to treatment-related mortality, with nonfatal toxicity not reported due to the retrospective nature of the study.

In summary, we present findings from a large retrospective analysis of upfront R-Mono or R-CHOP for B-cell PTLD arising after solid organ transplant. With selection of initial therapy made at the discretion of treating physicians, we report response and survival outcomes that compare favorably with those achieved with prior studies, albeit most patients had undergone renal or liver transplantation. Furthermore, we demonstrate that R-Mono and R-CHOP are equally efficacious for patients with low-risk PTLD. Meanwhile, although R-Mono confers inferior response in high-risk disease, this does not negatively impact survival because patients can subsequently receive R-CHOP. Overall, our results support the use of R-Mono as the default upfront therapy for PTLD, although given that R-CHOP was not associated with an excess of TRM, we contend that upfront R-CHOP remains a useful option for selected high-risk patients in whom a rapid response is desirable.

REFERENCES

1. Parker A, Bowles K, Bradley JA, et al.; Haemato-oncology Task Force of the British Committee for Standards in Haematology and British Transplantation Society. Diagnosis of post-transplant lymphoproliferative disorder in solid organ transplant recipients - BCSH and BTS guidelines. Br J Haematol. 2010; 149:675–692
2. Dierickx D, Habermann TM. Post-transplantation lymphoproliferative disorders in adults. N Engl J Med. 2018; 378:549–562
3. Rickinson AB, Kieff E. Fields Virology. 2007; Volume 2; 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins
4. Swerdlow S, Campo E, Harris NL, et al.; The International Agency for Research on Cancer. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press. 2016
5. Oertel SH, Verschuuren E, Reinke P, et al. Effect of anti-CD 20 antibody rituximab in patients with post-transplant lymphoproliferative disorder (PTLD). Am J Transplant. 2005; 5:2901–2906
6. Blaes AH, Peterson BA, Bartlett N, et al. Rituximab therapy is effective for posttransplant lymphoproliferative disorders after solid organ transplantation: results of a phase II trial. Cancer. 2005; 104:1661–1667
7. Choquet S, Leblond V, Herbrecht R, et al. Efficacy and safety of rituximab in B-cell post-transplantation lymphoproliferative disorders: results of a prospective multicenter phase 2 study. Blood. 2006; 107:3053–3057
8. González-Barca E, Domingo-Domenech E, Capote FJ, et al.; GEL/TAMO (Grupo Español de Linfomas); GELCAB (Grupo para el Estudio de los Linfomas Catalano-Balear); GOTEL (Grupo Oncológico para el Tratamiento y Estudio de los Linfomas). Prospective phase II trial of extended treatment with rituximab in patients with B-cell post-transplant lymphoproliferative disease. Haematologica. 2007; 92:1489–1494
9. Jain AB, Marcos A, Pokharna R, et al. Rituximab (chimeric anti-CD20 antibody) for posttransplant lymphoproliferative disorder after solid organ transplantation in adults: long-term experience from a single center. Transplantation. 2005; 80:1692–1698
10. Elstrom RL, Andreadis C, Aqui NA, et al. Treatment of PTLD with rituximab or chemotherapy. Am J Transplant. 2006; 6:569–576
11. Taylor AL, Bowles KM, Callaghan CJ, et al. Anthracycline-based chemotherapy as first-line treatment in adults with malignant posttransplant lymphoproliferative disorder after solid organ transplantation. Transplantation. 2006; 82:375–381
12. Buadi FK, Heyman MR, Gocke CD, et al. Treatment and outcomes of post-transplant lymphoproliferative disease: a single institution study. Am J Hematol. 2007; 82:208–214
13. Choquet S, Trappe R, Leblond V, et al. CHOP-21 for the treatment of post-transplant lymphoproliferative disorders (PTLD) following solid organ transplantation. Haematologica. 2007; 92:273–274
14. Trappe R, Oertel S, Leblond V, et al.; German PTLD Study Group; European PTLD Network. Sequential treatment with rituximab followed by CHOP chemotherapy in adult B-cell post-transplant lymphoproliferative disorder (PTLD): the prospective international multicentre phase 2 PTLD-1 trial. Lancet Oncol. 2012; 13:196–206
15. Trappe RU, Dierickx D, Zimmermann H, et al. Response to rituximab induction is a predictive marker in B-cell post-transplant lymphoproliferative disorder and allows successful stratification into rituximab or R-CHOP consolidation in an international, prospective, multicenter phase II trial. J Clin Oncol. 2017; 35:536–543
16. Leblond V, Dhedin N, Mamzer Bruneel MF, et al. Identification of prognostic factors in 61 patients with posttransplantation lymphoproliferative disorders. J Clin Oncol. 2001; 19:772–778
17. Ghobrial IM, Habermann TM, Maurer MJ, et al. Prognostic analysis for survival in adult solid organ transplant recipients with post-transplantation lymphoproliferative disorders. J Clin Oncol. 2005; 23:7574–7582
18. Choquet S, Oertel S, LeBlond V, et al. Rituximab in the management of post-transplantation lymphoproliferative disorder after solid organ transplantation: proceed with caution. Ann Hematol. 2007; 86:599–607
19. Hourigan MJ, Doecke J, Mollee PN, et al. A new prognosticator for post-transplant lymphoproliferative disorders after renal transplantation. Br J Haematol. 2008; 141:904–907
20. International Non-Hodgkin’s Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin’s lymphoma. N Engl J Med. 1993; 329:987–994
21. Dierickx D, Tousseyn T, Sagaert X, et al. Single-center analysis of biopsy-confirmed posttransplant lymphoproliferative disorder: incidence, clinicopathological characteristics and prognostic factors. Leuk Lymphoma. 2013; 54:2433–2440
22. Trappe RU, Choquet S, Dierickx D, et al.; German PTLD Study Group and the European PTLD Network. International prognostic index, type of transplant and response to rituximab are key parameters to tailor treatment in adults with CD20-positive B cell PTLD: clues from the PTLD-1 trial. Am J Transplant. 2015; 15:1091–1100
23. Dierickx D, Tousseyn T, Requilé A, et al. The accuracy of positron emission tomography in the detection of posttransplant lymphoproliferative disorder. Haematologica. 2013; 98:771–775
24. Zimmermann H, Denecke T, Dreyling MH, et al. End-of-treatment positron emission tomography after uniform first-line therapy of B-cell posttransplant lymphoproliferative disorder identifies patients at low risk of relapse in the prospective German PTLD registry. Transplantation. 2018; 102:868–875
25. Van Keerberghen CA, Goffin K, Vergote V, et al. Role of interim and end of treatment positron emission tomography for response assessment and prediction of relapse in posttransplant lymphoproliferative disorder. Acta Oncol. 2019; 58:1041–1047
26. Carbone PP, Kaplan HS, Musshoff K, et al. Report of the committee on Hodgkin’s disease staging classification. Cancer Res. 1971; 31:1860–1861
27. Lister TA, Crowther D, Sutcliffe SB, et al. Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin’s disease: Cotswolds meeting. J Clin Oncol. 1989; 7:1630–1636
28. Rohatiner A, d’Amore F, Coiffier B, et al. Report on a workshop convened to discuss the pathological and staging classifications of gastrointestinal tract lymphoma. Ann Oncol. 1994; 5:397–400
29. Cheson BD, Fisher RI, Barrington SF, et al.; Alliance, Australasian Leukaemia and Lymphoma Group; Eastern Cooperative Oncology Group; European Mantle Cell Lymphoma Consortium; Italian Lymphoma Foundation; European Organisation for Research; Treatment of Cancer/Dutch Hemato-Oncology Group; Grupo Español de Médula Ósea; German High-Grade Lymphoma Study Group; German Hodgkin’s Study Group; Japanese Lymphorra Study Group; Lymphoma Study Association; NCIC Clinical Trials Group; Nordic Lymphoma Study Group; Southwest Oncology Group; United Kingdom National Cancer Research Institute. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol. 2014; 32:3059–3068
30. Evens AM, David KA, Helenowski I, et al. Multicenter analysis of 80 solid organ transplantation recipients with post-transplantation lymphoproliferative disease: outcomes and prognostic factors in the modern era. J Clin Oncol. 2010; 28:1038–1046

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.