Direct Glycoprotein-Specific Platelet Autoantibodies Detected in a Majority of Patients with Chemotherapy-Induced Thrombocytopenia

Introduction: Chemotherapy-induced thrombocytopenia (CIT) is a common complication of cancer therapy attributed to myelosuppressive or cytotoxic effects of chemotherapy. CIT has been associated with increased mortality due to bleeding complications as well as reduced relative dose intensity as a result of dose reduction and/or treatment delay. The mechanism of CIT is believed to be due to myelosuppression and depletion of hematopoietic progenitors with repetitive, ongoing cytotoxic insults. However, both malignancy and cytotoxic chemotherapy result in immune dysregulation. To our knowledge, no study has rigorously evaluated for the presence of direct glycoprotein-specific platelet autoantibodies in patients with CIT. Methods: We performed an observational study of patients age ≥18 years with solid tumor malignancies diagnosed with CIT between 2018-2022 who had direct glycoprotein-specific platelet autoantibody (PA) testing performed as part of our institution's routine thrombocytopenia intake evaluation. PA testing was performed at a single central laboratory (Versiti Wisconsin) using a direct solid-phase ELISA assay detecting platelet bound antibodies binding to GPIIb/IIIa, GPIb/IX, and GPIa/IIa. PA testing was considered positive if optical density (OD) values were ≥ 2 times the mean of the negative control for each glycoprotein-specific assay. Positive values were scored as an ordinal variable representing integer multiples of the mean of the negative control up to a maximum of 5. We compared this data with PA testing results utilizing the same assay in a consecutive cohort of 96 patients at Versiti with positive PA testing for evaluation of presumed immune thrombocytopenia (ITP). The mean number of positive antibodies was compared using two sample t-test. Median glycoprotein-specific PA OD scores were compared between groups using the Wilcoxon rank sum test. Prevalence of positive antibody testing was compared between groups using one-way ANOVA. Relationships between number of positive PA, platelet nadir, and thrombopoietin level was assessed using Spearman's correlation. Results: 55 patients with CIT were included representing over ten tumor types (pancreatic, colorectal, and lung were most common) with a median age of 64 years old. The median platelet nadir (within one month of PA testing and prior to TPO-RA initiation) was 50x109/L (IQR=33-63). 36 of 55 patients (65.4%) with CIT had positive PA testing. Within the PA positive CIT patients, the mean number of positive glycoprotein-specific tests was 2.2; testing for GPIIb/IIIa, Ib/IX, and Ia/IIa, specifically, was positive in 63.9%, 61.1%, and 97.2%, respectively. The mean number of positive antibodies was not significantly different between PA positive CIT patients and the control cohort (2.22 vs 2.51, respectively. p=0.0823). When comparing positive PA OD values in our PA positive CIT cohort compared with those of the control cohort, there was no significant difference for GPIIb/IIIa, Ib/IX, and Ia/IIa. We observed a substantially higher rate of positive results for GPIa/IIa in PA positive CIT patients compared with figures previously reported in ITP patients with positive PA testing (97.2% vs 51.8%, Al-Samkari et al. Blood Adv 2020). The number of positive PAs was not significantly correlated with either platelet nadir (rs=-0.1426, p=0.2990) or thrombopoietin level (rs=0.2044, p=0.1462). Thrombopoietin level was significantly negatively correlated with platelet nadir (rs=-.4037, p=0.0030). 8 patients with CIT (14.6%) had received immune checkpoint inhibitors within 6 months of PA testing and there was no significant difference in PA positivity in these patients compared with the 47 patients who had not received immune checkpoint inhibitors. Conclusion: We observed a striking rate of positive results for direct platelet autoantibody testing in patients with CIT-a population for which very low rates would be expected- using a modern, high-fidelity, glycoprotein-specific direct PA assay. The magnitude of PA optical densities in CIT were comparable to those of a control cohort being evaluated for likely ITP signifying equivalent laboratory assay significance. These results are consistent with known immune dysregulation in cancer patients receiving cytotoxic chemotherapy and raise provocative questions regarding a potential immune-mediated mechanism of thrombocytopenia in many patients with CIT.