Combined Elevated Anti-Adamts-13 Antibody Level and Low Adamts-13 Activity during Remission Highly Predicts Disease Relapse in a Prospective Cohort of Patients with Autoimmune Thrombotic Thrombocytopenic Purpura (TTP)

Introduction

Autoimmune TTP (aTTP) is a rare life-threatening microangiopathy characterized by hemolytic anemia, thrombocytopenia, and severe ADAMTS-13 deficiency. Daily therapeutic plasma exchange (TPE) and preemptive use of rituximab have considerably improved patient outcome. However, a high proportion of patients still experience relapse and several studies are evaluating the role of ADAMTS-13 activity and its inhibitors to predict disease recurrence. In this study, in a large prospective cohort of aTTP patients, we aimed to characterize the therapeutic approach, the clinical outcome and the predictive variables of relapse.

Methods

From 2002 to 2018, 163 TTP cases were referred to our Centers. Blood samples were obtained from all patients at diagnosis. In 84 patients (58F/26M) samples were collected also during follow-up: 74 had autoimmune, 4 congenital, 3 drug-induced, and 3 BMT-related TTP. ADAMTS-13 activity and inhibitors by mixing studies (chromogenic assay), and anti-ADAMTS-13 Ab (ELISA) were measured in plasma. A complete ADAMTS-13 activity deficiency (<10%) was diagnostic for TTP. An anti-ADAMTS-13 Ab titer >15U/ml was considered positive. Hemoglobin (Hb), platelets (Plt), and hemolysis parameters (i.e. aptoglobin (apt), bilirubin, LDH, reticulocytes) were monitored during follow up. Treatment response was defined as: complete response (CR) = Plt>150x10⁹/L for at least 2 days and LDH normalization, lasting for 30 days; exacerbation = recurrent disease within 30 days after reaching CR; refractoriness = no treatment response by day 30. Data are expressed as median and range.

Results

All 74 patients with aTTP had complete ADAMTS-13 activity deficiency (0%, range 0-9) and high anti-ADAMTS-13 Ab (70 U/ml, 22-373). In acute phase, blood parameters were: Plt 2x10⁹/L (12-49), Hb 8.8 g/dL (4.1-12.8), apt 5.5 mg/dL (0.1-44), and LDH 1688 U/L (312-7007). All patients were treated in first-line with TPE and high-dose corticosteroids (median n. TPE 9, 5-24). Vincristine, high-dose immunoglobulin and N-acetylcysteine were used in 9, 4, and 1 cases, respectively. Fifty-eight patients were evaluable for treatment response. Two early deaths were due to myocardial infarction and diffuse rectal ischemia after 7 and 10 days from diagnosis, respectively (unresponsive to 7 and 8 cycles of TPE). Twenty-five patients (43%) obtained a CR after a median time to response (mTTR) of 10 days (4-18) and a median number (n.) of TPE of 7 (5-14); 19 obtained a suboptimal response: 12 exacerbation (mTTR 31days (16-62), median TPE n. 14 (7-24)), and 7 refractory (mTTR 42.5 (27-86), median TPE n. 11 (8-19)). ADAMTS-13 activity, Ab, Pts, Hb and hemolysis parameters at diagnosis were not predictive for suboptimal response. Sixteen patients with a suboptimal response were treated with Rituximab, obtaining a CR in all cases.

At remission, 55% of patients showed normal levels (>50%) of ADAMTS-13, while 20% moderate (21-50%), 8.6% severe (10-20%), and 12% complete (<10%) deficiency of ADAMTS-13 activity. The normalization or the partial recovery of ADAMTS-13 activity was associated to negative Ab levels in 91% of cases.

Relapse occurred in 30 patients (42%), of whom 17 had multiple relapses (median 4, 2-9). Median follow-up was 3.8 years (0-24). Thirty-three patients with complete clinical records and a follow-up ≥6 months were evaluated for predictive parameters of disease recurrence. Laboratory analysis showed that a low ADAMTS-13 activity (≤20%) during remission did not significantly predict for relapse (p=0.09); however 6/9 patients with anti-ADAMTS-13 Ab level ≥20 U/L during remission had at least 1 relapse, as compared to 5/20 patients with Ab < 20 U/L (p=0.03). Importantly, the combination of Ab levels ≥20 U/L and ADAMTS-13 activity <20% strongly predicted for relapse (p=0.0004). Conversely, hematological and hemolysis parameters did not correlate with relapse. Of note, patients with a suboptimal response treated with preemptive rituximab in addition to first-line TPE did not relapse.

Conclusions

In conclusion, our study shows that serial measurements of ADAMTS-13 activity and anti- ADAMTS-13 Ab are warranted. Indeed, the combination of the two variables was highly predictive of relapse in our prospective series of patients. Further studies are needed to validate this approach to identify aTTP patients at high risk of relapse and guide the decision on early intervention.