Citations

Abstract Background Despite advances in the diagnosis and treatment of immune thrombotic thrombocytopenic purpura (iTTP), recent mortality estimates remain around 5–10%. Predicting iTTP mortality risk has been of interest, as it may help individualize treatment or potentially guide the use of more intensive or novel therapies based on disease severity. However, prior attempts at mortality prediction have yielded inconsistent results, and no individual variable has demonstrated independent or reliable predictive ability. In a previous analysis utilizing the USTMA iTTP registry, which included 419 patients from 15 TTP referral centers, we reported that mortality due to immune TTP was unpredictable (Blood. 2021;138[Suppl 1]:1047). Neither the French Thrombotic Microangiopathy (TMA) Reference Score (previously validated in the French TMA Registry), nor a previous logistic regression model attempt (developed by USTMA investigators in 2021) demonstrated sufficient accuracy in predicting iTTP mortality in the USTMA dataset (AUC = 0.63 and 0.69, respectively). We subsequently utilized gradient boosted machine, a machine learning approach capable of handling multiple continuous variables and the complex interactions between them in large datasets, and developed a novel prediction tool: the USTMA Mortality Index (Res Pract Thromb Haemost. 2024; 8(3):102388). It demonstrated good performance in predicting iTTP mortality (AUC = 0.77) in repeated 10-fold cross-validation. However, it has not yet undergone external validation. Objectives To externally validate the USTMA Mortality Index to predict mortality due to acute iTTP in a multicenter cohort. Methods We performed an external validation using an independent cohort of iTTP patients from 5 institutions: University of Utah, Case Western Reserve University, Rochester Regional Health, University of Illinois, and Mayo Clinic. Patients in the validation cohort were not part of the development cohort (the USTMA registry). We included patients with confirmed iTTP and defined acute iTTP mortality as death within 30 days of admission. USTMA Mortality Index variables were collected at initial presentation, which included (in order of highest to lowest relative influence): lactate dehydrogenase, serum creatinine, age, hemoglobin, platelet count, presence of stupor/coma, fever. The model's prediction accuracy was assessed using area under the receiver operating characteristic curve (AUC). We calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) based on the previously defined cut-offs of 4.27% and 9.08% mortality risk in the original model, which categorized patients into standard risk, high risk, and very high risk of predicted mortality. Results The external validation cohort included 103 patients with confirmed iTTP and 11 deaths (10.7% mortality rate). The USTMA Mortality Index demonstrated poor performance with AUC of 0.69 (95% CI, 0.55–0.83) in the validation cohort. The wide confidence interval reflects substantial uncertainty in model performance, ranging from poor to excellent, and is likely due to the small number of deaths. At a risk threshold ≥4.27%:Sensitivity = 64%Specificity = 70%PPV = 20%NPV = 94% At a risk threshold ≥9.08%:Sensitivity = 18%Specificity = 87%PPV = 14%NPV = 90% Conclusion The USTMA Mortality Index demonstrated poor performance in the external validation cohort (AUC = 0.69, 95% CI, 0.55–0.83), contrasting with its previously favorable performance in repeated 10-fold cross-validation. This discrepancy likely reflects model overfitting, a known challenge in AI-based clinical modeling. Our study is limited by the small number of deaths, resulting in a wide confidence interval ranging from poor to excellent performance. Given that neither the French TMA Reference Score, nor our group's prior logistic regression model (2021), nor the USTMA Mortality Index (2024) could reliably predict mortality upon external validation in our cohorts, these findings collectively suggest that mortality in acute iTTP remains difficult to predict at this time, regardless of statistical methodology or current machine learning capabilities. Clinical decision-making, including the use of therapies such as caplacizumab, should not be guided solely by mortality risk predictions. Instead, efforts to reduce mortality and improve outcomes should prioritize individualized clinical assessment over model-based stratification at this time.

Abstract Background Due to the high mortality of immune thrombotic thrombocytopenic purpura (iTTP), rapid clinical evaluation and early intervention are crucial. The PLASMIC score remains a valuable and widely used tool for predicting severe ADAMTS13 deficiency. However, recent studies have shown variable performance across different populations. In addition, its variables may often be missing during initial presentation. Refining the accuracy of early iTTP evaluation could improve timely recognition and treatment, while reducing unnecessary interventions in those eventually ruled out for iTTP. Machine learning advancements offer several advantages for enhancing clinical prediction models: the ability to assess continuous variables, capture complex interactions between them, and handle missing data. We previously utilized XGBoost (extreme gradient boosting) to develop a novel machine learning tool, TTP-14, which improved upon the PLASMIC score's predictive performance in repeat 10-fold cross-validation achieving an AUC of 0.91 (Blood. 2024;144[Suppl 1]:2617). TTP-14 generates a predicted probability for iTTP, expressed as a continuous percentage estimate of iTTP diagnostic likelihood, and is also able to do so with missing variables. However, it has not previously been externally validated. Objectives To externally validate the ability of TTP-14 to predict iTTP diagnosis and improve upon the PLASMIC score. Methods We performed an external validation of TTP-14 using an independent cohort of patients with suspected iTTP at the Mayo Clinic. These patients were not part of the original multi-center development cohort of 106 iTTP patients from the University of Utah, Case Western Reserve University, Rochester Regional Health, and University of Illinois. For the validation cohort, we identified patients who underwent ADAMTS13 activity testing between 2010 and 2025 for evaluation of thrombotic microangiopathy at the Mayo Clinic. iTTP was defined as ADAMTS13 activity <10%, and controls as ADAMTS13 activity >20%. TTP-14 variables were collected at initial presentation: platelet count, lactate dehydrogenase, creatinine, international normalized ratio, reticulocyte percentage, active cancer, indirect bilirubin, mean corpuscular volume, total bilirubin, haptoglobin, neurologic deficit, hemoglobin, age, and history of transplant. Receiver operating characteristic curve (AUC) was reported with 95% confidence intervals (CI). While the model provides a continuous estimate and is not categorical, we evaluated clinically meaningful probability thresholds to illustrate performance in decision-relevant zones. Sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were reported at these select thresholds and analyzed for patients with and without missing variables. Result We included 50 iTTP patients and 56 controls in the external validation cohort. 13 patients had missing PLASMIC variables. TTP-14 achieved an AUC of 0.96 (95% CI: 0.93–0.99), outperforming the PLASMIC score's AUC of 0.90 (95% CI: 0.83–0.96) in the same dataset, and improving upon TTP-14's previously reported cross-validated AUC of 0.91 (95% CI: 0.90–0.92) in the development cohort. Among patients with complete data (n = 93), applying select thresholds for clinical relevance demonstrated: Low risk cut-off (<5% predicted iTTP risk): Sensitivity = 98% Specificity = 71% PPV = 76% NPV = 97% Accuracy = 84% High risk cut-off (≥50% predicted iTTP risk): Sensitivity = 71% Specificity = 98% PPV = 97% NPV = 78% Accuracy = 85% In the full cohort (n = 106) that included those with missing PLASMIC variables, TTP-14 maintained excellent predictive performance with AUC = 0.95 (95% CI: 0.92–0.99). Performance metrics analyzed using the above thresholds also yielded similar results. Conclusion External validation of TTP-14 demonstrated high predictive accuracy for iTTP diagnosis (AUC = 0.96). TTP-14 outperformed the PLASMIC score in the validation cohort (AUC = 0.96 vs 0.90, respectively). TTP-14 offers more flexibility than the PLASMIC score by generating predictions even with incomplete data, and maintained excellent performance in the cohort that included those with missing variables (AUC = 0.95) Its refined accuracy and ability to predict iTTP despite missing data provide a key advantage, offering an alternative tool to improve rapid iTTP diagnosis. TTP-14 may help enhance early recognition and treatment of iTTP while reducing unnecessary interventions in non-TTP patients.

Background: Vaccine-induced immune thrombotic thrombocytopenia (VITT) is a severe thrombotic thrombocytopenia that predominantly occurs after adenoviral vector-based vaccines. VITT patients produce anti-platelet factor 4 (PF4) antibodies with restricted clonality (i.e., mono/oligoclonal) with lambda light chains. Long-term follow-up studies in VITT show a disconnect in the persistence of anti-PF4 antibodies, whereby ELISA-positive antibodies persist significantly longer than platelet-activating antibodies. The mechanistic basis of this is unclear and could be explained by a couple of hypotheses: (1) Epitope spreading where, with time, the specificity of platelet-activating anti-PF4 antibodies may “spread” to recognize only PF4 complexed to heparin vs. (2) Persistence of clonally restricted platelet-activating anti-PF4 antibodies which decrease in titers with time such that their levels are insufficient to crosslink platelet FcGRIIa and induce platelet activation. Methods: Acute, convalescent patient sera and affinity-purified anti-PF4 antibodies were evaluated in antigen-based, functional (platelet-activation-based), and mass spectrometry studies as previously described (Blood, 2022 Jul 7;140(1):73-77). The median time to last follow-up was 243 days after the acute episode (range: 114-664 days). Results: Follow-up samples were obtained from 6 VITT patients after Ad26.COV2.S vaccination. At last follow-up, anti-PF4 IgG antibodies reactive in PF4/polyanion ELISA were detected in the sera of 5/6 VITT patients. However, only one of these samples stimulated platelet activation in the PF4-dependent P-selectin expression assay (PEA; 199 days after acute presentation). Affinity-purified anti-PF4 antibodies were detected by ELISA in all follow-up samples and three of the six affinity-purified anti-PF4 antibody samples activated platelets in the PEA. Four of the six VITT patients demonstrated a monoclonal anti-PF4 antibody, while one each had a biclonal and triclonal profile. Five VITT patients produced IgG lambda light chain antibodies, while one produced an IgG kappa. All six VITT patients demonstrated synonymous anti-PF4 antibody profiles when comparing acute and follow-up antibody repertoires, without evidence of VITT patients producing new anti-PF4 antibodies after the acute episode. In all cases, levels of anti-PF4 antibodies were significantly lower in convalescent samples relative to acute samples. Given this data on the persistence of clonally-restricted VITT antibodies, samples from another VITT patient with thrombocytopenia (100-150,000/uL) persisting for more than three years were similarly evaluated. This patient, who experienced severe thrombosis after ChAdOx1 nCoV19 vaccination, demonstrated persistent ELISA and PEA positivity even three years after symptom onset but had not experienced breakthrough thrombosis while on anticoagulation therapy. Serum protein electrophoresis (SPEP) did not detect a monoclonal gammopathy. Anti-PF4 antibodies immuno-enriched from a sample obtained >2 years after acute presentation demonstrated a monoclonal profile. While non-availability of an acute sample for anti-PF4 antibody characterization precluded a firm determination of whether this “Long VITT” case represented a subset of monoclonal gammopathy of thrombotic significance (MGTS; Blood 2023 Apr 6;141(14):1772-1776; New Eng J Med 2024 Aug 8), the chronicity of thrombocytopenia, presence of a monoclonal anti-PF4 antibody and persistently positive anti-PF4 serology supports this diagnosis. Conclusions. Clonally-restricted VITT antibodies persist for months-years with no evidence for the production of novel anti-PF4 antibodies after the acute event. Decreased abundance of anti-PF4 antibodies leads to negative functional results with follow up samples. Some platelet-activating antibodies can persist for very long periods (>3 years) and mediate persistent thrombocytopenia, representing an SPEP-negative MGTS-like state. This study also demonstrates the first example of a VITT antibody with a kappa light chain. The requirement for anticoagulation in patients demonstrating ELISA-positive/functional test-negative results is not clear, but the persistence of mono/oligoclonal anti-PF4 antibodies raises the possibility for the upregulation of pathogenic antibody production in proinflammatory settings.

AbstractHeparin-induced thrombocytopenia (HIT) is suspected much more often than it is confirmed. Technically simple platelet factor 4 (PF4)-polyanion enzyme-linked immunosorbent assays (ELISAs) are sensitive but nonspecific. In contrast, accurate functional tests such as the serotonin release assay, heparin-induced platelet activation assay, and PF4-dependent P-selectin expression assay require fresh platelets and have complex assay end points, limiting their availability to specialized reference laboratories. To enable broad deployment of functional testing, we sought to extend platelet viability significantly by optimizing storage conditions and developed a simple functional assay end point by measuring the release of a platelet α-granule protein, thrombospondin-1 (TSP1), in an ELISA format. Platelet cryopreservation conditions were optimized by freezing platelets at controlled cooling rates that preserve activatability. Several-month-old cryopreserved platelets were treated with PF4 or heparin and were evaluated for their ability to be activated by HIT and vaccine-induced immune thrombotic thrombocytopenia (VITT) antibodies in the TSP1 release assay (TRA). HIT and spontaneous HIT patient samples induced significantly higher TSP1 release using both PF4-treated (PF4-TRA) and heparin-treated cryopreserved platelets relative to samples from patients suspected of HIT who lacked platelet-activating antibodies. This latter group included several patients that tested strongly positive in PF4-polyanion ELISA but were not platelet-activating. Four VITT patient samples tested in the TRA activated PF4-treated, but not heparin-treated, cryopreserved platelets, consistent with recent data suggesting the requirement for PF4-treated platelets for VITT antibody detection. These findings have the potential to transform the testing paradigm in HIT and VITT, making decentralized, technically simple functional testing available for rapid and accurate in-hospital diagnosis.

Kanack and colleagues analyze anti-platelet factor 4 antibodies from 5 patients with vaccine-induced thrombotic thrombocytopenia (VITT) secondary to COVID-19 adenoviral vaccination and antibodies from patients with spontaneous heparin-induced thrombocytopenia (HIT) and classical HIT. VITT antibodies are monoclonal or oligoclonal, similar to spontaneous HIT, whereas classical HIT antibodies are polyclonal. Heparin inhibits antibody-induced platelet activation in VITT, suggesting that heparin should be considered for the treatment of VITT.

AbstractRare cases of COVID‐19 vaccinated individuals develop anti‐platelet factor 4 (PF4) antibodies that cause thrombocytopenia and thrombotic complications, a syndrome referred to as vaccine‐induced immune thrombotic thrombocytopenia (VITT). Currently, information on the characteristics and persistence of anti‐PF4 antibodies that cause VITT after Ad26.COV2.S vaccination is limited, and available diagnostic assays fail to differentiate Ad26.COV2.S and ChAdOx1 nCoV‐19‐associated VITT from similar clinical disorders, namely heparin‐induced thrombocytopenia (HIT) and spontaneous HIT. Here we demonstrate that while Ad26.COV2.S‐associated VITT patients are uniformly strongly positive in PF4‐polyanion enzyme‐linked immunosorbent assays (ELISAs); they are frequently negative in the serotonin release assay (SRA). The PF4‐dependent p‐selectin expression assay (PEA) that uses platelets treated with PF4 rather than heparin consistently diagnosed Ad26.COV2.S‐associated VITT. Most Ad26.COV2.S‐associated VITT antibodies persisted for >5 months in PF4‐polyanion ELISAs, while the PEA became negative earlier. Two patients had otherwise unexplained mild persistent thrombocytopenia (140‐150 x 103/µL) 6 months after acute presentation. From an epidemiological perspective, differentiating VITT from spontaneous HIT, another entity that develops in the absence of proximate heparin exposure, and HIT is important, but currently available PF4‐polyanion ELISAs and functional assay are non‐specific and detect all three conditions. Here, we report that a novel un‐complexed PF4 ELISA specifically differentiates VITT, secondary to both Ad26.COV2.S and ChAdOx1 nCoV‐19, from both spontaneous HIT, HIT and commonly‐encountered HIT‐suspected patients who are PF4/polyanion ELISA‐positive but negative in functional assays. In summary, Ad26.COV2.S‐associated VITT antibodies are persistent, and the un‐complexed PF4 ELISA appears to be both sensitive and specific for VITT diagnosis.

Abstract Background: Platelet-activating anti-PF4 antibodies that cause thrombocytopenia and thrombosis are implicated in Heparin-induced thrombocytopenia (HIT) and, more recently, in a newly recognized entity, vaccine-induced immune thrombotic thrombocytopenia (VITT). Current frontline PF4-Polyanion ELISAs utilize simple techniques but are highly non-specific. Functional gold-standard assays such as the serotonin-release assay (SRA) and PF4-dependent P-selectin expression assay (PEA) have limited availability due to technical complexity and the requirement for freshly-drawn normal donor platelets which results in significant delays in diagnosis. Thus, there is an unmet need for near-patient, functional assays for HIT/VITT diagnosis. Aim: To develop a highly sensitive and specific functional assay to detect platelet-activating antibodies associated with HIT and VITT using long-term stored platelets. The assay should be technically simple, avoiding current functional HIT testing endpoints such as aggregometry, radioactivity measurement or flow cytometry. Methods: Platelets were subjected to controlled-rate freezing in a trehalose-based buffer. Cryopreserved platelets were then thawed after variable storage periods, incubated with platelet factor 4 (PF4), and sera from HIT or VITT suspected patients. Thrombospondin-1 (TSP1), a protein highly expressed in platelet α-granules, was quantified from activated platelet supernatants in an ELISA-based assay as a measure of platelet activation. Results: In initial studies, p-selectin expression was induced by platelet-activating HIT antibodies using cryopreserved platelets and the TSP1-release assay (TRA) detected HIT antibody-mediated PF4-dependent cryopreserved platelet activation across 20 different platelet lots (data not shown). A representative study using platelets stored for 11 months is shown in Figure 1A. A cohort of thirty-six HIT-suspected patient sera, half with PF4-dependent platelet-activating antibodies (PEA+) and without (PEA-), were used to evaluate the diagnostic accuracy of the TRA (Figure 1B). All 18 PEA+ samples activated platelets strongly with a mean TSP1 concentration of 147 µg/mL measured in the supernatant compared to a mean of 43 µg/mL obtained with non-activating samples. TSP1 release in a subgroup of 7 PEA+/SRA+/ELISA+ and 4 PEA-/SRA-/ELISA+ samples that were matched for ELISA Optical Density was examined (Figure 1C). Results demonstrated that even strong ELISA+ samples that were non-activating (PEA-/SRA-) did not induce platelet activation in the TRA, confirming high specificity of the assay. The TRA was tested for its ability to detect platelet-activating antibodies in four patients who developed VITT after the Janssen vaccine (Ad26.COV2.S). All samples induced high levels of TSP1 release from cryopreserved platelets (Figure 1D). Conclusions: Our results demonstrate that cryopreserved platelets stored for several months are viable and capable of degranulation upon activation. Further, we demonstrate that measurement of TSP-1 released from cryopreserved, PF4-treated platelets is highly accurate for detecting platelet-activating HIT and VITT anti-PF4 antibodies. Specificity and sensitivity were both 100% even with samples that were strongly false-positive in the HIT ELISA. Coupling cryopreserved platelet activation to a simple ELISA endpoint, as demonstrated here, precludes the need for fresh platelets and complex testing techniques. Thus, the TRA has the potential to transform the diagnostic testing paradigm in HIT and VITT by making near-patient in-hospital functional testing available for rapid and accurate diagnosis. Figure Legend: Figure 1. TSP-1 quantification of (A) 3 HIT patient samples (closed circles) relative to 3 normal control samples (open circles) after 11-months of frozen platelet storage. (B) Thirty-six HIT-suspected patient samples: 18 PEA+ samples (closed circles), and 18 PEA- samples (open squares). Data are compared using two-tailed, unpaired Student's t test. **** P<0.0001. (C) Eleven high optical density matched HIT ELISA patient sera: 7 PEA+/SRA+ samples (closed circles) and 4 PEA-/SRA- samples (open squares). (D) TSP-1 quantification of VITT patient sera. NC-normal control. Figure 1 Figure 1. Disclosures Pruthi: Instrumentation Laboratory: Honoraria; HEMA Biologics: Honoraria; Bayer Healthcare AG: Honoraria; Genentech: Honoraria; CSL Behring: Honoraria; Merck: Honoraria. Padmanabhan: Veralox Therapeutics: Membership on an entity's Board of Directors or advisory committees.