Plasma Exchange Taper for Acquired TTP Is Protective Against Recurrence at Both 30 Days and 6 Months: A Retrospective Study from 2 Academic Medical Centers

Background: Acquired Thrombotic Thrombocytopenic Purpura (TTP) is a hematologic disorder characterized by microangiopathic hemolytic anemia and thrombocytopenia, frequently accompanied by ADAMTS13 deficiency from inhibiting and/or clearing antibodies. The current standard of care is to emergently institute daily therapeutic plasma exchanges (TPE) and immunosuppression until a treatment response is achieved; however, there is clinical equipoise as to whether or not to taper TPE. No studies to date have directly compared these strategies. We hypothesized that tapering plasma exchange would protect against recurrence at 30 days but likely only delay recurrence, and thus by 180 days recurrence rates would not differ.

Methods: Subjects were identified from previously established acquired TTP registries at two academic medical centers: one where TPE is nearly universally tapered and one where TPE is never tapered. Inclusion criteria included documentation of either ADAMTS13 activity < 10% or < 15% with an inhibitor. For each patient, episodes were excluded if the patient received agents other than steroids or rituximab (including splenectomy) during the episode or prophylaxis therapy of any kind. Non-tapered patients from the tapering institution were excluded to prevent selection bias. For each TTP episode, immunosuppression therapy, time to disease recurrence or death, and central venous catheter infections from each center were recorded. In order to control for the effect of immunosuppression, prednisone-treated episodes were analyzed separately from rituximab-treated episodes. Kaplan-Meier curves were created for each unique group and Log-rank test was used to compare them. Categorical variables were compared with Fisher's exact test. Statistical significance was defined as p<0.05.

Results: Overall, 46 unique patients were tapered vs. 61 who were not. For prednisone-only treated episodes, 52 were tapered and 57 were not. At 30 days, 46 of 52 tapered episodes (88%) treated with prednisone were free of recurrence versus 31 of 56 episodes (54%) without a TPE taper; a comparison of Kaplan-Meier curves demonstrated a statistically significant difference in recurrence-free survival (p<0.0001); HR=4.1 (95% CI, 2.0-8.2) for the non-tapered strategy. Recurrence-free survival at 180 days was 71% (37/52) for episodes treated with prednisone and TPE taper versus 46% (26/57) without a taper. A comparison of Kaplan-Meier curves demonstrated a statistically significant difference that persists at 180 days (p<0.01); HR=2.5 (95% CI, 1.4-4.6).

For rituximab-treated episodes, 21 were tapered and 38 were not. At 30 days, 20 of 21 tapered episodes (95.2%) treated with rituximab were free of recurrence versus 31 of 38 episodes (81.6%) treated with rituximab without a plasma exchange taper; a comparison of Kaplan-Meier curves demonstrated a trend towards statistical significance in recurrence-free survival (p=0.10); HR=3.1 (95%CI, 0.8-12.1) for the non-tapered strategy. Recurrence-free survival at 180 days was 90% (19/21) for episodes treated with rituximab and plasma exchange taper versus 76% (29/38) treated with rituximab without a taper (p=0.18); HR 2.3 (95%CI, 0.7-7.9).

There were significantly more central venous catheter infections at the tapering institution (12 of 46) compared to the non-tapering institution (3 of 61; p=0.004).

Conclusions: In this retrospective analysis, we found that after a treatment response using prednisone and TPE, there was a highly significant difference in recurrence rate at 30 days without a TPE taper, which was an expected finding. However, at 180 days, the difference between the groups persisted, which suggests that the effect of the taper was not simply to delaying the recurrence, but, rather, contributes to obtaining a durable treatment response. We hypothesize that this effect had contributions from a delayed effectiveness of prednisone therapy and an immune-tolerance effect. There were no significant findings in the rituximab-treated episodes, but there was trend toward a similar finding of a significant difference at 30 days. This may suggest that the effect of rituximab is faster than prednisone and thus the benefit of tapering TPE when treating with rituximab is less clear, but we were limited by small samples sizes. Catheter-related infections were also higher in the taper group, indicating that the taper should be used parsimoniously.