Plasma levels of S100A8/A9, histone/DNA complexes, and cell-free DNA predict adverse outcomes of immune thrombotic thrombocytopenic purpura

Abstract

Background:

Immune thrombotic thrombocytopenic purpura (iTTP) is a life-threatening blood disorder, primarily resulting from autoantibodies against ADAMTS13. Infection or inflammation often precedes acute iTTP. However, the association of inflammation and inflammatory mediators with disease severity and outcome of acute iTTP is not fully assessed.

Objectives:

Here, we determined plasma levels of S100A8/A9, histone/DNA complexes, citrullinated histone H3 (CitH3), and cell-free DNA (cfDNA) in a cohort of 108 acute episodes from 94 unique iTTP patients and healthy controls, and assessed the association of each of these biomarkers with the disease severity and mortality.

Results:

All acute iTTP patients had significantly increased plasma levels of S100A8/A9 (median 84.8, interquartile range [IQR] 31.2–157.4 μg/mL), histone/DNA complexes (median 55.7, IQR 35.8–130.8 U/mL), CitH3 (median 3.8, IQR 2.2–6.4 ng/mL), and cfDNA (median 937.7, IQR 781.3–1420.0 ng/mL) on the admission blood samples when compared with healthy controls. An increased plasma level of S100A8/A9, histone/DNA complex and cfDNA was associated with organ damage, coagulopathy, and mortality in iTTP. After being adjusted for age and history of hypertension, Cox proportional hazard regression analysis demonstrated that a hazard ratio (95% confidence interval) for an elevated plasma level of S100A8/A9, histone/DNA complexes, and cfDNA was 11.5 (1.4–90.9) (P = .021), 10.3 (2.7–38.5) (P = .001), and 12.8 (3.9–42.0) (P = .014), respectively.

Conclusion:

These results indicate that inflammation or plasma inflammatory mediators such as S100A8/A9 or NETosis markers such as histone/DNA complexes and cfDNA may play a role in pathogenesis of iTTP, which may help stratify patients with a high risk of death during acute iTTP episodes.

1 |. INTRODUCTION

Thrombotic thrombocytopenic purpura (TTP) is a rare, but life-threatening thrombotic disorder. It occurs in 3 to 10 cases per million per year.^1,2 Most cases of TTP are caused by autoantibodies against ADAMTS13,³ a plasma metalloprotease that cleaves von Willebrand factor (VWF), referred to as immune-mediated TTP (iTTP). Rarely, TTP may be caused by hereditary deficiency of ADAMTS13 activity, resulting from mutations of ADAMTS13,⁴ referred to as hereditary or congenital TTP (cTTP). Regardless of its etiology, severe deficiency of plasma ADAMTS13 is necessary, but often not sufficient to result in an acute episode of TTP. This was demonstrated in patients with cTTP⁵ or in a murine model of TTP,^6–10 suggesting that additional triggers may be needed for the development of a full-blown TTP episode.

Previous studies have demonstrated that infections may result in acute inflammation and activation of neutrophils, which leads to the release of neutrophil azurophilic granular contents, including S100A8/A9,^11,12 human neutrophil peptides 1–3,¹³ and neutrophil extracellular traps (NETs).^14,15

S100A8 and S100A9, also known as calgranulins A and B or myeloid-related proteins (MRP) 8 and 14. S100A8 and S100A9 form stable heterodimers and are primarily expressed in neutrophils, monocytes,¹⁶ dendritic cells,¹⁷ and mature macrophages.^18,19 Plasma S100A8/A9 heterodimers are primarily released from activated or necrotic neutrophils and monocytes/macrophages, the innate immune mediators involving in pathogenesis of various inflammatory diseases. Elevated S100A8/A9 were shown to be a useful diagnostic marker for inflammation, especially in noninfectious conditions such as arthritis and chronic inflammatory lung²⁰ and bowel disease.²¹

Cell-free DNA (cfDNA) is released through various cellular processes, including apoptosis, necrosis, or by neutrophils, as a component of NETs.^22,23 NETs are composed of neutrophil proteases and histone/DNA complexes. These mesh-like structures may not only ensnare and eliminate microbes, but also provide scaffolding for platelet adhesion,²⁴ activated factor VII and tissue factor,²⁵ and inhibit tissue factor pathway inhibitor, antithrombin, and protein C pathway,^25,26 thus increasing the potential for thrombus formation.

Plasma S100A8/A9, cfDNA, and histone/DNA complexes are found to be dramatically increased in patients with acute iTTP.^14,27,28 Moreover, the DNase I, which degrades DNA components, was shown to be significantly reduced in patients with acute thrombotic microangiopathy.²⁹ However, the clinical relevance of these inflammatory mediators or NETosis markers to disease severity and outcome of iTTP is not fully appraised. The present study aims to determine the association of plasma S100A8/A9, cfDNA, and histone/DNA complexes with the severity and outcome of patients with acute iTTP after standard of care consisting of therapeutic plasma exchange (TPE), corti-costeriods, and rituximab.

2 |. PATIENTS AND METHODS

2.1 |. Patients

Ninety-four patients with 108 acute episodes of iTTP admitted to the University of Alabama at Birmingham Medical Center from April 2006 to June 2019 were enrolled into the study. This cohort of patients was of course previously reported for other biomarker analyses.^14,30 Control samples were collected from healthy donors who were age-, gender-, and ethnic background-matched, but did not have a history of hematological diseases, malignancy, and acute inflammatory disorders. The institutional review boards (IRB) at the University of Alabama at Birmingham (UAB) and University of Kansas Medical Center (KUMC) have approved the study protocol.

Demographic information, routine laboratory results, and treatments were collected from the electronic medical records. Additional follow-ups were conducted through outpatient visits, telephone interviews, and interviews at locally organized annual TTP Fairs. Diagnostic criteria for iTTP were the same as previously described.^14,30,31 These include severe thrombocytopenia, microangiopathic hemolytic anemia with or without end-organ damage; and plasma ADAMTS13 activity <10 IU/dL with a detectable ADAMTS13 inhibitor or anti-ADAMTS13 IgG. Myocardial involvement was defined as troponin I above the normal range (ie, >0.03 ng/mL) on admission; neurological involvement was defined using a reduced

Glasgow Coma Scale (GCS) as described previously.³² Clinical response is defined as having achieved a sustained normalization of platelet counts above the lower limit of the established reference range (eg, >150 × 10⁹/L) and of lactate dehydrogenase (LDH; <1.5 × upper limit of normal) for 2 consecutive days after the cessation of plasma exchange; clinical remission is defined as having achieved the clinical response for at least 30 days after the cessation of TPE.^31,33

2.2 |. Blood collection

Whole blood samples were collected from informed and consented patients before the initiation of TPE and anticoagulated with 3.2% sodium citrate. Additional blood samples were collected from 46 patients at clinical response or during remission. The blood samples were centrifuged at 1,500g for 15 minutes and plasma was aspirated and stored in aliquots in a −80°C freezer until assays.

2.3 |. Assay for plasma S100A8/A9

Plasma levels of S100A8/A9 were determined using a human S100A8/A9 DuoSet ELISA kit (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. The linearity of this assay was from 0 to 6.0 ng/mL with a coefficient of variation of <10%.

2.4 |. Assays for plasma histone/DNA complexes and CitH3

Histone/DNA complexes (Millipore Sigma) and citrullinated histone H3 (Cayman Chemicals, Ann Arbor, MI) were determined by the ELISA methods according to the manufacturer’s instructions. Arbitrary units equal to the ratio of optical density of a patient plasma sample to the optical density of a pooled human plasma.

2.5 |. Assay for plasma cfDNA

Quant-IT PicoGreen dsDNA assay (Thermo Fisher Scientific) was used to quantify plasma levels of cfDNA according to the manufacturer’s instructions. Linear detection was 0 to 1.0 ng/mL.

2.6 |. Assay for plasma ADAMTS13 activity and inhibitor

Plasma ADAMTS13 activity and inhibitors were determined in the reference laboratory (Versiti, Milwaukee, WI) or an in-house FRETSVWF73 assay as previously described.⁶

2.7 |. Anti-ADAMTS13 IgG

Plasma anti-ADAMTS13 IgG was determined by an ELISA as previously described.^14,30

2.8 |. Statistical analysis

All statistical analyses were performed with the SPSS 25.0 software. For data with normal distribution, values were expressed as the mean ± standard deviation (SD). For data that were not normally distributed, the values were expressed as the median and interquartile range (IQR). Student t test was used to determine differences between two groups with normally distributed data, while the Mann Whitney U test was used to analyze data that were not normally distributed. The Wilcoxon test was used for paired data that were not normally distributed; the Fisher exact test was used for categorical variables. Receiver operating characteristic curves were used to identify the optimal cutoff point for each of these inflammatory markers to predict the mortality. Furthermore, the log-rank test was used to compare the death-free survival rate between two groups and the Cox proportional hazard regression was used to determine the hazard ratios for the predictive variables. P values of <.05 and <.01 were considered statistically significant and highly statistically different, respectively.

3 |. RESULTS

3.1 |. Patient characteristics

Of 108 acute episodes (94 unique iTTP patients), the median age was 46.5 (IQR, 34.3–54.8) years. Of 94 patients, 53 (56.4%) patients were female and 75 (79.8%) were African American. Of 108 acute episodes, 65 (60.2%) were initial episodes and 43 (39.8%) relapsed. The overall episode mortality rate was 10.2% (11/108) with a median time of death of 10 (IQR, 2–14) days from admission. There was no significant difference in demographic parameters, comorbidities, and clinical presentation except for the GCS between the patients who died and those who survived (Table 1).

TABLE 1.

Demographic information, comorbidities, and presenting symptoms in 108 iTTP episodes from 94 unique patients who died vs survived

Variables	Survived (N = 97)	Died (N = 11)	P Value
Demographic
Age, y*	45 (34.0–52.5)	54 (49.0–56.0)	.052a
Female (N = 94), n (%)	47 (56.0)	6 (60.0)	1.000
African American (N = 94), n (%)	68 (81.0)	7 (70.0)	.421
Disease status, initial, n (%)	60 (61.9)	5 (45.5)	.340
Blood group O (N = 94), n (%)	53 (54.6)	5 (45.5)	.751
Comorbidities
Hypertension, n (%)	48 (49.5)	8 (72.7	.097
Diabetes mellitus, n (%)	22 (22.7)	2 (18.2)	1.000
SLE
Positive, n (%)	8 (8.2)	0 (0)	.320
Negative, n (%)	36 (37.1)	0 (0)	.013
ND, n (%)	53 (54.6)	11 (100)	.004
HIV
Positive, n (%)	6 (6.2)	0 (0)	.400
Negative, n (%)	61 (62.9)	5 (45.5)	.260
ND, n (%)	30 (30.9)	6 (54.5)	.120
Symptoms
CNS symptoms, n (%)	52 (53.6)	8 (72.7)	.339
GCS	15 (14–15)	10 (3–15)	<.001
Chest pain, n (%)	12 (12.4)	1 (9.1)	.751
Abdominal pain, n (%)	25 (25.8)	4 (36.4)	.751

Note: Age and GCS were expressed as the median and interquartile range, but other parameters were expressed as the number and the percentage. Boldface type indicates that the difference is statistically significant.

Abbreviations: CNS, central nervous system; GCS, Glasgow Coma Scale; HIV, human immunodeficiency virus; n, number of cases/episodes; ND, not determined or not available; SLE, systemic lupus erythematosus.

^aMann-Whitney U test; all other analyses were performed using χ2 test.

3.2 |. Plasma levels of S100A8/A9, histone/DNA complexes, CitH3, and cfDNA in iTTP patients and healthy controls

S100A8/A9 is mainly derived from immunocytes, such as neutrophils and macrophages, which participate in the inflammatory process.^34–36 Histone/DNA, CitH3, and cfDNA are released from NETosis, a regulated process of neutrophil cell death that contributes to the host defense against pathogens.^37,38 As shown, the admission plasma levels of S100A8/A9, histone/DNA, CitH3, and cfDNA in patients with acute iTTP were dramatically increased compared with those in the healthy controls (Figure 1). The median (IQR) plasma levels of S100A8/A9, histone/DNA complexes, CitH3, and cfDNA in patients with acute iTTP were 84.8 (31.2–157.4) μg/mL, 55.7 (35.8130.8) U/mL, 3.9 (2.2–7.1) ng/mL, and 937.7 (781.3–1420.4) ng/mL, respectively. Interestingly, patients who were admitted with an initial iTTP episode appeared to have higher plasma levels of S100A8/ A9, histone-DNA complexes, and cfDNA, but not CitH3 when compared with those with the relapsed episodes (Figure S1).

FIGURE 1.

Plasma S100A8/A9, histone/DNA complexes, CitH3, and cfDNA in patients with acute iTTP and healthy controls. Plasma levels of (A) S100A8/A9, (B) histone/DNA complexes, (C) CitH3, and (D) cfDNA in patients with acute iTTP on admission and in healthy controls. All data points are presented as individual values (dots), median, and interquartile range (IQR) (horizontal bars). Mann-Whitney U test was performed to determine the statistical significance. ****P value <.0001

As expected, these inflammatory mediators or NETs markers were dramatically reduced in iTTP patients at clinical response/remission with only a few exceptions (Figure 2). Of 46 patients who were assessed during clinical response, seven still had low ADAMTS13 activity. However, there was no statistically significant difference in these biomarkers in patients with or without severe ADAMTS13 deficiency (Figure S1). These results are consistent with those reported previously,^14,15 further supporting the notion that acute iTTP is an inflammatory autoimmune disorder, which persists after cessation of TPE and normalization of platelet counts.

FIGURE 2.

Plasma S100A8/A9, histone/DNA complexes, CitH3, and cfDNA in iTTP patients during acute episodes and at clinical response/remission. Plasma levels of (A) S100A8/A9, (B) histone/DNA complexes, (C) CitH3, and (D) cfDNA in iTTP patients on admission and at clinical response/remission. Wilcoxon test was performed to determine the statistical significance between the two time-points. *, ***, **** indicate P value <.05, .001, and <.0001, respectively

3.3 |. Correlations between each of various inflammatory or NET markers and organ damage

To determine the relationship between inflammatory or NET markers and end-organ damage, Spearman correlation coefficients were determined between each of the inflammatory markers and each of the organ damage markers such as LDH and troponin-I. As shown, plasma levels of S100A8/A9 (r = 0.46, P < .0001), CitH3 (r = 0.24, P = .0015), histone/DNA complexes (r = 0.53, P < .0001), and cfDNA (r = 0.35, P = .0002) were all positively correlated with LDH (Figure S2); additionally, plasma levels of S100A8/A9, histone/DNA complexes, and cfDNA were positively correlated with prothrombin time, activated partial thromboplastin time, D-dimer, and plasminogen activator inhibitor-1 levels (data not shown), which are critical parameters for coagulopathy. Moreover, patients with elevated levels of serum troponin-I exhibited significantly higher plasma levels of S100A8/A9, histone/DNA complexes, and cfDNA, but not CitH3 (Figure 3). Interestingly, patients with a low GCS had significantly higher plasma levels of histone/DNA complexes and cfDNA, but not S100A8/9 and CitH3 when compared with those who had normal GCS (Figure S3). These results indicate that plasma S100A8/A9 may be associated with myocardial damage, but plasma histone/DNA complexes and cfDNA appear to be associated with both myocardial and neurological damage. Interestingly, plasma levels of S100A8/9, histone/DNA, complexes, CitH3, and cfDNA were not associated with plasma ADAMTS13 activity during remission (Figure S4), suggesting that these markers are independent variables for assessing the disease process of iTTP.

FIGURE 3.

Plasma S100A8/A9, histone-DNA complexes, CitH3, and cfDNA in iTTP patients with or without myocardial ischemia. Plasma levels of (A) S100A8/A9, (B) histone/DNA complexes, (C) cfDNA, and (D) CitH3 in iTTP patients with and without myocardial involvement. Eighty patient episodes showed elevated troponin levels, but 26 patients had normal troponin. However, one patient with normal troponin did not have sufficient plasma for assessing plasma CitH3. All data points are expressed as individual values (dots), median, and interquartile range (IQR) (horizontal bars). Mann-Whitney U test determined the statistical significance. ns, **, and *** indicate P value >.05, <.01, and <.005, respectively

3.4 |. Increased plasma levels of inflammatory or NETs markers were associated with mortality in acute iTTP

The mortality for acute iTTP in this cohort of patients was 10.2%, similar to what we have previously reported.¹⁴ Demographic features of 11 patients who died and 83 patients (with 97 episodes) who survived are summarized in Table 1. There was no statistically significant difference in the demographic characteristics except for the GCS between the two groups. Patients who died exhibited significantly higher admission levels of serum LDH, activated partial thromboplastin time, plasma fibrinogen, D-dimers, and troponin-I than those who survived the acute episodes (Table 2). Not previously recognized or reported, patients who died also had significantly higher admission median levels of plasma S100A8/A9 (198.2 vs 88.1 μg/mL) (P = .024), histone/DNA complexes (79.6 vs 55.3 U/ mL (P = .019), and cfDNA (9132.0 vs 4287.8 ng/mL) (P < .001) than those who survived (Table 2).

TABLE 2.

Routine and special laboratory parameters in 108 iTTP episodes on admission associated with death and survival

Variables	Survived (n = 97)	Died (n = 11)	P Value
WBC, ×109/L	10.3 (8.5–13.5)*	18.9 (7.7–23.8)	.052a
Hb, g/dL	8.6 ± 2.0**	7.8 ± 2.6	.275b
Hct, %	24.9 ± 5.6	22.6 ± 7.4	.340b
Platelet count, ×109/L*	13 (9.1–22)	11.5 (7.5–29.2)	.437a
LDH, U/L	1006 (688–1621)	1545.5 (925–5937)	.016a
LDH/Hb ratio	133.3 (72.2–216.1)	265.5 (120.6–738.9)	.017a
Creatinine, mg/dL	1.2 (0.9–1.7)	3.5 (1.5–5.5)	.076a
PT, s	14.7 (14.1–15.7)	16.6 (14.5–19.9)	.036a
aPTT, s	30 (27–34)	35.5 (29.3–57.5)	.024a
Fibrinogen, mg/dL	424.4 ± 122.4	557.5 ± 178.7	.006b
D-dimer, ng/mL	1869 (1112.0–4468.0)	7083.0 (2165.8–10 810)	.036a
Troponin-I, ng/mL	0.11 (0.03–0.77)	0.47 (0.23–14.4)	.047a
ADAMTS13 activity, U/dL	2.4 (0.2–4.9)	0 (0–1.4)	.345a
ADAMTS13 antigen, ng/mL	55.0 (27.2–119.3)	94.8 (43.7–170.1)	.532a
Anti-ADAMTS13 IgG, U/ mL	4385.6 (2711.9–8187.1)	4567.2 (1435.8–9890.3)	.606a
ADAMTS13 inhibitor, U/mL	1.4 (0.8–3.2)	3.2 (1.5–6.2)	.177a
S100A8/A9, μg/mL	88.1 (31.4–163.7)	198.2 (38.0–251.2)	.024a
CitH3, ng/mL	3.9 (2.3–6.9)	2.9 (2.3–7.4)	.535a
Histone/DNA complexes, U/mL	55.3 (35.8–128.3)	79.6 (44.0–178.6)	.019a
cfDNA, ng/mL	913.0 (790.5–1381.6)	4287.8 (1237.0–6775.9)	<.001a

Note: Hb, Hct, and Fibrinogen are expressed as mean and standard and deviation; other parameters are expressed as mean and interquartile range. Boldface type indicates that the difference is statistically significant.

Abbreviations: aPTT, activated partial thromboplastin time; cfDNA, cell-free DNA; CitH3, citrullinated histone 3; Hb, hemoglobin; Hct, hematocrit; LDH, lactate dehydrogenase; n, number of cases; PT, prothrombin time; WBC, white blood cell.

^aMann Whitney U.

^bt test. Normal range for troponin-I is <0.03 ng/mL, determined by Beckman Coulter DXI 800 analyzer. The data are presented either the

^*mean ± standard deviation (SD) or the

^**median (interquartile range, IQR).

Using receiver operating characteristic curves, we were able to identify optimal cutoff points for each inflammatory marker to predict mortality (Figure S5): 102.6 μg/mL for S100A8/A9 with a sensitivity of and a specificity of 90.9% and 62.5%, 176.2 U/mL for histone/DNA complexes with a sensitivity of and a specificity of 54.5% and 89.7%, 2.2 ng/mL for CitH3 with a sensitivity of and a specificity of 74.0% and 74.0%, and 1097.3 ng/mL for cfDNA with a sensitivity of and a specificity of 90.9% and 66.7%, respectively. Using these cutoff values, we determined the survival rates over time between patients who had higher and lower levels of these biomarkers using Kaplan-Meier survival analysis. As shown, patients with plasma levels of S100A8/A9 > 102.6 μg/mL (P = .0006), histone/DNA complexes >176.2 U/mL (P < .001), CitH3 > 2.2 ng/mL, and cfDNA > 1097.3 ng/mL (P = .001) had significantly lower survival rates than those with values below these cutoffs (Figure 4).

FIGURE 4.

Kaplan-Meier survival analysis of iTTP patients with high and low plasma levels of S100A8/A9, histone-DNA complexes, CitH3, and cfDNA. Survival probability (%) predicted in patients with high (red line) and low (blue line) levels of S100A8/A9, histone/DNA complexes, CitH3, and cfDNA in iTTP patients as indicated in each figure. Log-rank test was performed to obtain the statistical significance. P values <.05 and <.01 are considered to be statistically significant and highly significant, respectively

Moreover, Cox proportional hazard regression analysis demonstrated that after adjusting for confounding factors (eg, age, history of hypertension), patients with higher plasma levels of S100A8/A9, histone/DNA complexes, and cfDNA had 11.5 (95% confidence interval [CI] 1.4–90.9, P = .021), 10.3 (2.7–38.5, P = .001), and 12.8 (3.9–42.0, P = .014) times higher risk of disease exacerbation, respectively, than those with values below these cutoffs. In this analysis, we also found that patients with low GCS (≤14, hazard ratio [HR] 7.1, 95% CI 1.8–27.8), elevated serum LDH (>1682 U/L, HR 6.6, 95% CI 1.8–24.4), and an increased LDH/hemoglobin ratio (>254, HR 9.7, 95% CI 2.6–37.0) also had an increased risk of exacerbation (Figure 5).

FIGURE 5.

Cox proportional hazard regression analysis for identification of laboratory markers associated with iTTP exacerbation. The cut off of the left was determined by receiver operating characteristic curve and Forest plot in the middle demonstrates the relative risk of exacerbation (defined by the disease recurs within 30 d of stopping TPE). Hb, hemoglobin. Covariates are age and history of hypertension. HR, hazard ratio; 95% CI, 95% confidence interval. P values <.05 and .01 are considered to be statistically significant and highly significant, respectively

Together, all these results demonstrate that admission plasma levels of S100A8/A9, histone/DNA complexes and cfDNA, as well as markers of hemolysis and organ damage (ie, neurological involvement) are predictive for disease severity, in-hospital mortality, and recurrence (exacerbation or relapse) after achieving a clinical response in patients with acute iTTP.

4 |. DISCUSSION

The present study demonstrates the crucial role of plasma inflammatory and NETs markers in predicting the severity, in-hospital mortality, and the rate of exacerbation of acute iTTP despite intensive therapy with TPE, steroids, and rituximab, etc. The mortality rate remains at ~10% in this cohort, as we previously reported.^14,39,40

Previous studies have suggested that old age,⁴¹ high levels of LDH activity,⁴¹ elevated levels of serum troponin-I,^14,42 neurological involvement,⁴¹ low ADAMTS13 antigen,^43,44 and high anti-ADAMTS13 IgG⁴³ at presentation may be associated with poor outcomes. Additionally, we and others have demonstrated that elevated plasma levels of soluble terminal complement attack complexes (ie, sC5b-9) may also be predictive for mortality.^14,45

All patients with acute iTTP have dramatically increased levels of plasma S100A8/A9, histone/DNA complexes, CitH3, and cfDNA on admission; the levels of these biomarkers are dramatically reduced when patients are in clinical response/remission. The elevated plasma levels of histone/DNA complexes or cfDNA are primarily the results of neutrophil activation and death, a process known as NETosis,^46,47 but may also be contributed by the reduction of plasma DNase I activity in acute iTTP patients.²⁹ Interestingly, there was a significant association between each of the inflammatory or NET markers and each of the other laboratory parameters suggestive of organ damage or coagulopathy in acute iTTP. Most importantly, high admission levels of plasma S100A8/A9, histone-DNA complexes, citH3, and cfDNA are significantly associated with the in-hospital mortality of these patients.

Upon infection or during inflammatory responses, a cytosolic protein, arginine deiminase-4, is activated, resulting in citrullination of histones that leads to NETosis,³⁴ a unique form of cell death characterized by the release of decondensed chromatin into the extracellular space. cfDNA and histone/DNA complexes are the primary structural components of NETs. Plasma NET components are dramatically increased in patients with cancer,⁴⁸ ischemic stroke,⁴⁹ sepsis,⁵⁰ and autoimmune disease,⁵¹ as well as immune thrombocytopenia.²⁴

NETs may activate platelets and provide scaffold for red blood cells, fibrin, and VWF.⁵¹ Also, the specific NETs components such as free histone, cfDNA, and histone/DNA complexes may activate the intrinsic and extrinsic coagulation cascade pathways and platelets while reducing tissue factor pathway inhibitor and protein C activity, leading to enhanced potential for thrombus formation.^34,46 Recent studies demonstrate that purified histone can directly trigger endothelial exocytosis, resulting in massive release of VWF from Weibel-Palade bodies and thus thrombocytopenia in mice⁵² and triggering a “TTP-like” syndrome in ADAMTS13-deficient zebrafish.⁹

S100A8/A9, a heterodimer of two calcium-binding proteins, was originally discovered as immunogenic protein expressed and secreted by neutrophils. Subsequently, it emerges as an important proinflammatory mediator in acute and chronic inflammation. More recently, it is demonstrated that S100A8/A9 may play a role in thrombosis. In vitro thrombus formation is significantly reduced in whole blood from Mrp14 (ie, S100A9)^−/− mice.⁵³ S100A8 and S100A9 are also known as myeloid-related protein-8 (Mrp8) and myeloid-related protein-14 (Mrp14), respectively.^54,55 Platelet-expressed S100A8/ A9 accelerates the development of artery occlusion after injury, suggesting that platelet-derived S100A8/A9 may directly promote thrombosis.⁵³

Although this is the first comprehensive appraisal of the predictive role of inflammatory or NETosis markers in iTTP, there are some limitations to our study. First, the sample size is still relatively small because only a few patients died during acute episodes. A multicenter study may help address such a limitation; second, our patient sampling spans from 2006 to 2019. In the past decade, more aggressive or perhaps more novel therapeutic approaches, such as upfront use of rituximab^56,57 and early addition of caplacuzimab,^39,58 may have circumvented the usual prognostic factors of iTTP if the new ISTH guidelines are fully adopted.^59,60 However, this patient cohort did not include patients who were treated with caplacizumab. The treatment protocol at our institutions has been relatively unchanged (eg, daily TPE and steroids, with rituximab in 40% of patients). The wide adoption of ISTH guidelines for the diagnosis and management of TTP in many parts of the world remains a slow and uneven process; therefore, having some prognostic markers for patient stratification may still be useful as some clinicians are still reluctant of using caplacizumab upfront.

In summary, we found that plasma inflammatory danger signals (eg, S100A8/A9) and NET biomarkers (eg, histone-DNA complexes, citH3, and cfDNA) are dramatically increased in patients with acute iTTP; these elevated plasma biomarkers are predictive of the severity, in-hospital mortality, and on-treatment exacerbation or recurrence of iTTP. Our findings, along with other clinical and laboratory parameters, may help clinicians stratify their patients for a more intensive management strategy, including intensive care unit admission, upfront caplacizumab,^58,61 and careful monitoring for vital signs. This may lead to further reduction of in-hopsital mortality and subsequent recurrence of iTTP.

Essentials.

• Biomarkers predicting outcomes of immune thrombotic thrombocytopenic purpura (iTTP) are still lacking.

• Plasma changes of S100A8/A9, histone/DNA, and cell-free DNA in iTTP patients were determined.

• Plasma S100A8/A9, histone/DNA, and cell-free DNA were dramatically increased in acute iTTP.

• The admission plasma levels of S100A8/A9, histone/DNA, and cell-free DNA predict mortality in acute iTTP.