Title: Pretreatment ADAMTS13 Activity Predicts Clinical Outcome in Patients with Acquired Thrombotic Thrombocytopenic Purpura (TTP)

Acquired TTP is characterized by a significant risk for exacerbation (recurrent TTP within 30 days of last plasma exchange (PEX)). An algorithm applied early in the course of treatment that detects patients at increased risk for exacerbation could identify patients most likely to benefit from more intensive upfront therapy (PEX, immune-suppressive therapy) to prevent recurrence of disease after tapering PEX. We have previously reported in a small cohort study that pretreatment ADAMTS13 activity was significantly lower (0.8% v. 1.4%) in patients suffering an exacerbation, but the difference was primarily mediated by race (EJH 2009 Dec 1;83(6):559–64). We performed a study to analyze data from a large cohort of TTP patients to assess the potential prognostic utility of pretreatment ADAMTS13 activity for risk of exacerbation.

Patients presenting with a clinical diagnosis of acquired TTP (platelets < 100K/uL, microangiopathic hemolytic anemia) between May 2002 and August 2010 with pretreatment ADAMTS13 activity data (86 acute TTP episodes from 49 patients evaluable for exacerbation, and 94 episodes for other factors associated with ADAMTS13 activity) were included in these analyses.

Demographic and laboratory data obtained prior to starting PEX in patients with acute TTP were analyzed retrospectively to determine their relationship to early recurrences (exacerbation) of TTP. Based on these data, ROC analysis and recursive partitioning algorithms were used to determine an optimal cutoff of pretreatment ADAMTS13 activity that was prognostic for exacerbation of TTP. Subsequently, the acute TTP episodes with pretreatment ADAMTS13 <1.15% were compared to the cohort with ADAMTS13 >1.15% with respect to clinical outcomes and presenting clinical symptoms and laboratory data. Impact of these factors on exacerbation were also analyzed using generalized estimating equation models to account for multiple episodes from the same patients.

Overall, there were 22 exacerbation events observed in the 86 TTP episodes. There were no significant differences between subjects that suffered an exacerbation in terms of clinical laboratory data (platelet count, LDH, creatinine) at presentation. However, mean ADAMTS13 activity at presentation was significantly lower in patients that suffered exacerbations compared to those that did not [1.4% (0.5 –7.2) v. 2.9% (0.5–13.3); p=0.002]. ROC analysis of all study subjects indicated that the optimal threshold for pretreatment ADAMTS13 activity to predict the risk of exacerbation is 1.15%, which provides a test sensitivity of 64% and specificity of 62%. Subset analyses also defined the optimal threshold for AA and female populations to predict an increased risk of relapse to be 2% for both groups, giving rise to a sensitivity of 100% (AA)/82%(F) and specificity of 42% (AA)/50% (F), respectively. When using an ADAMTS13 activity threshold of 1.15% to risk stratify for increased risk for exacerbation, there were no significant differences in presenting clinical symptoms and laboratory values (Table 1). However, subjects with pretreatment ADAMTS13 activity <1.15% were more likely to suffer an exacerbation and require a greater number of PEX procedures to achieve a clinical response (platelets >150 K/uL).

These data are among the first to suggest pretreatment ADAMTS13 activity may predict clinical outcome from an acute episode of TTP. The optimal ADAMTS13 activity threshold was 1.15%, with ADAMTS13 activity below this level being associated with an increased risk for exacerbation. Furthermore, incorporating this threshold of ADAMTS13 activity into a model with race and gender will significantly improve risk stratification and identify patients at greatest risk for early treatment failure who may be more likely to benefit from the early addition of immune-based therapy to PEX.

No relevant conflicts of interest to declare.