Retrospective Analysis of Lymphomas in the Setting of Autoimmune Disease and the Impact of Immunosuppression

 

Introduction: Post-transplant lymphoproliferative disease (PTLD) encompasses a heterogeneous array of cases of lymphoma/lymphoma-like conditions arising in the setting of immunosuppression (IS) for prior organ or marrow transplant.  Such pts face heightened risk of toxicity from exposure to cytotoxic chemotherapy, and may be best treated in the frontline with reduction of IS (RI) and anti-CD20 monoclonal antibody treatment (Trappe, 2012).  Autoimmune (AI) disease has been associated with an increased risk of developing lymphoma; however, the relative impact of baseline clinical features, including prior IS, is unknown.  

Methods: We conducted a multicenter, retrospective analysis of adult pts with pre-existing AI conditions who were diagnosed with lymphoma since 1997. Baseline clinical features at diagnosis of lymphoid malignancy, including International Prognostic Index (IPI) risk factors; underlying AI disease; duration and type of IS; EBV status (by EBER in-situ hybridization); and primary therapy received (RI, rituximab [R] monotherapy, chemotherapy [+/- R]); were collected. Survival analyses were performed using Kaplan-Meier method.  We then focused on those who had A) received IS other than corticosteroids (CS) alone; and B) those diagnosed with DLBCL. Those variables found to have significant correlation with OS by univariate analyses (UVA) were used to construct Cox proportional hazards model (multivariate analysis [MVA]) in order to determine which might have the strongest association with OS. Lastly, we sought to evaluate a potential role for RI and/or R as frontline therapy for those with DLBCL.

Results: A total of 130 pts were included (Table 1). The most frequent AI disease was rheumatoid arthritis and for all cases, 76% had documented exposure to IS, for a median duration of 4.5 years (range 0.17-57 years) prior to diagnosis of lymphoma. The most common histologic subtype was DLBCL (52%).  EBV status was reported for only 34% of pts, but was positive in 68% (25/37), all of whom had received prior exposure to IS beyond CS, and 80% of whom (20/25) were diagnosed with DLBCL.  EBV status was infrequently tested in pts not previously exposed to IS (3/31).  At a median follow-up of 61 months for the entire cohort, 2-year PFS and OS were 79% and 91%, respectively (Figure 1, Panel A). By UVA, age>60; PS>1; LDH> upper limit of normal (ULN); DLBCL (vs all other histologies); underlying rheumatoid arthritis (RA; vs all other AI diseases); and prior exposure to IS, each correlated with inferior OS (Table 1). By MVA, PS>1 and prior IS maintained significance (p<0.05). If those receiving only CS are grouped with those not previously exposed to IS, the correlation of this factor with OS was strengthened (p 0.008), and by MVA, PS>1 (p 0.002) and prior IS (p 0.010) maintain significance (data not shown).

Among 67 pts with DLBCL, median age was 61 (range 26-90), 60% had advanced stage disease, and 32% had IPI of 4 or 5. At a median follow-up of 32 months, the 2-year PFS and OS were 82% and 84%, respectively. There were no differences in frequency of any IPI factors between patients exposed to prior IS (n=53) and those who were naďve to prior IS (n=14).  For those not exposed to prior IS, the 2-year OS was 100%, compared to 80% in those who received  prior IS (p 0.24); corresponding 2-year PFS were 92% and 79%, respectively (p 0.41). Age>60 and PS>1 were associated with an inferior OS but use of IS was not associated with outcome (Table 2).  The 2 year OS  for those treated with R plus CHOP(like) chemotherapy, CHOP(like) chemotherapy (without R), R alone (+/- RI), and with RI alone were 92%, 75%, 90%, and 67%, respectively (Figure 1, Panel B; log-rank p value 0.55).  Patients who received CHOP-like therapy +/- R, as compared to R and/or RI were more likely to be naďve to IS therapy (15/46 vs 0/22, p = 0.003) and have 2 or more EN sites of disease (15/46 vs 2/22,  =0.041). These differences notwithstanding, the 2-year PFS for the two groups were 86% and 74% (p 0.16), and 2-year OS for the two groups were 88% and 82%, respectively (Figure 1, Panel C; p 0.91).

Conclusions: Pts with immunosuppression-related lymphoma have high rates of 2-year OS and in DLBCL, IS does not appear to be associated with an inferior outcome. Similar to evolving treatment paradigms in PTLD, rituximab monotherapy and other cytotoxic chemotherapy-free regimens as well as risk-adapted approaches may warrant further evaluation in IS-related DLBCL.