Characterization of platelet autoantibody prevalence, serology, and predictive value for progression to immune thrombocytopenia in patients with persistent isolated mild thrombocytopenia (PIMT)

Abstract

Abstract BACKGROUND Persistent isolated mild thrombocytopenia (PIMT) is a recently defined and soon-to-be formalized immune thrombocytopenia (ITP) prodromal state, defined as a platelet count between 100 and 149 × 109/L of unknown etiology. No guidelines exist to direct workup or surveillance of these patients and there is no data to guide prognostic discussion on the risks of development of more defined hematologic pathology, such as ITP or hematologic neoplasia. Without a comprehensive evaluation including molecular testing for acquired and inherited genetic mutations and bone marrow evaluation, diagnoses of acquired clonal disorders, hereditary thrombocytopenia syndromes, and idiopathic cytopenia of undetermined significance cannot be made or ruled out. Therefore, management may vary considerably from reassurance only to a comprehensive evaluation at the time of referral. Previous studies have demonstrated that patients with PIMT have a 1-2% chance per year of progression to frank hematologic disease, most commonly ITP. The aim of this study was to characterize the prevalence, serologic pattern, and clinical predictive significance of platelet autoantibodies (PA) in patients with PIMT. METHODS We performed an observational study evaluating the results of direct and indirect glycoprotein-specific PA testing in patients with PIMT. All data was collected via manual chart review by physicians. All PA testing was performed by Versiti/BloodCenter of Wisconsin following all ISTH PA testing recommendations. PA testing included the PakAuto (Immucor, Brookfield, WI) through 2018, or its successor assay, the Versiti Platelet Autoantibody Assay (Versiti, Milwaukee, WI) from 2018 to present. Both assays evaluate for presence of glycoprotein (GP)-specific antibodies (anti-GPIIb/IIIa, anti-GPIb/IX, or anti-GPIa/IIa) in plasma (indirect) and eluted off of the platelet surface (direct). RESULTS Patients. 144 patients with PIMT evaluated and diagnosed by a hematologist had direct and indirect glycoprotein-specific PA testing performed between 2015 and 2024. The mean age was 64 years (range, 21-96 years), 54% were female, and 21% were of non-white race. Prevalence and Serology: 100 out of 144 patients (69%) had positive PA testing on initial assay, including 88 patients (61%) with antibodies detected via elution from the platelet surface (direct testing) and 38 patients (26%) with antibodies detected in plasma (indirect testing). Anti-GP IIb/IIIa antibodies were detected in 77 patients (54%), including 72 (50%) on direct and 11 (8%) on indirect testing. Anti-GP Ib/IX antibodies were detected in 73 patients (51%), including 70 (49%) on direct and 8 (6%) on indirect testing. Anti-GP Ia/IIa antibodies were detected in 88 patients (61%), including 77 (54%) on direct and 28 (19%) on indirect testing. Of 88 patients positive for direct PA, 58 (66%) were positive for all 3 GP-specific PA, 16 (18%) were positive for 2 GP-specific PA, and 16 (18%) were positive for 1 GP-specific PA. Predictive Value for Progression to ITP: 41 of 144 patients (29%) developed hematologic disease over the observation period, including 28 patients (19%) who progressed to ITP and 13 patients (9%) who developed a clonal myeloid disorder (7 patients) or clonal lymphoid disorder (6 patients). Of the 28 patients who progressed to ITP, 26 (93%) had positive PA testing, compared with 66 of 103 patients (64%) who had not progressed to hematologic disease during the observation period (odds ratio, 7.29, 95% confidence interval 1.75-32.27, P=0.0023). CONCLUSIONS This study defines for the first time the prevalence, serology, and predictive value of glycoprotein-specific PAs in PIMT. PAs are common in PIMT, occurring in 69% of patients (compared with 90% in 228 patients with known ITP from the same institution undergoing the same PA testing). Anti-GPIIb/IIIa, anti-GPIb/IX, and anti-GPIa/IIa antibodies each occurred in approximately 50-60% of patients with PIMT. Given that patients with PIMT and positive PA testing had 7-fold greater odds of later developing ITP compared with patients with PIMT who did not have positive PA testing, PA testing may have predictive value in the evaluation of patients with PIMT.