Use of Preemptive Treatment for Immune Thrombotic Thrombocytopenic Purpura: A Matched Survival Analysis

Background:

Immune Thrombotic thrombocytopenic purpura (iTTP) is a relapsing disorder, resulting from depletion of ADAMTS13. With the goal of preventing clinical relapse there has been considerable interest in ADAMTS13 monitoring and preemptive treatment. Indeed, preemptive treatment with rituximab has shown significant promise and is now included in the 2020 International Society on Thrombosis and Haemostasis (ISTH) treatment guidelines. Evaluation of long-term outcomes of ADAMTS13 based preemptive treatment have generally relied on limited numbers of historical controls which did not incorporate matching for known risk factors for relapse such as race, number of prior episodes, and immunosuppression. Our study aims to leverage the data in the United States Thrombotic Microangiopathy (USTMA) iTTP registry to compare the clinical relapse-free survival (RFS) intervals between iTTP patients treated preemptively based on ADAMTS13 level and those treated only for clinical relapse base on clinical and laboratory manifestations of iTTP.

Methods:

We carried out a retrospective study utilizing the multi-center cohort of patients with iTTP from the USTMA Consortium (spanning 1985-2019 and containing 780 participant records). We selected for patients with relapsing disease treated either preemptively based on ADAMTS13 level or for clinical relapse. Each preemptively treated episode was matched to up to five clinically relapsed episodes using covariate balancing propensity score nearest-neighbor matching by age, gender, race, prior relapse status, acute treatment, treatment center, and calendar year, with exact matching required for treatment center and prior relapse status. Time from prior episode to relapse was compared between preemptively treated episodes and matched controls using a Cox proportional hazards model weighted by the matching weights, including subclasses as a cluster, and using cluster-robust standard errors.

Results:

We identified 1068 episodes of iTTP. Of these. Thirteen participants accounting for a total of seventeen preemptive treatment episodes were included in the data set however, we included only the thirteen first uses of preemptive treatment to avoid selection bias. These were matched with 59 episodes treated for clinical relapse. Over 70% of the included participants in both groups were identified as black/African American. Distribution of immunosuppressive medications used in clinically relapsed cases were not significantly different than those used in preemptive treatment, and corticosteroids were the most frequently used. Demographic and treatment data before and after matching are included in Table 1. There was no significant difference in RFS between the groups, hazard ratio 0.63 (95% confidence interval 0.26-1.54), p=0.31.

Conclusion:

To our knowledge this is the first study to use matching controls to obtain a causal estimate of the effect of ADAMTS13 monitoring based preemptive treatment on RFS in iTTP. In contrast to previous work, we did not observe a significant increase in RFS in the preemptively treated group. It is possible that the limited number of preemptively treated episodes herein did not provide sufficient power to detect a difference. However, given the significant effect size that has been proposed, based on prior work, we suspect other factors may have also influenced this finding. Previous analysis of the USTMA iTTP registry has shown a higher incidence of relapse in African American participants. While the precise reason for this finding is not yet clear, it is worth considering that there may also be differential response to preemptive treatment. Our study highlights the importance of continued efforts to determine the generalizability and efficacy of ADAMTS13 monitoring based preemptive treatment in iTTP.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal