Global Hemostatic Assays for Risk Stratification of Thrombosis in Myeloproliferative Neoplasms (MPN): A Systematic Review and Meta-Analysis

Background: Arterial and venous thromboembolism are major complications of myeloproliferative neoplasms (MPN), which include polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF). The pathogenesis of thrombosis in MPN is not fully understood. Global hemostatic assays, including thrombin generation assay (TGA), rotational thromboelastometry (ROTEM), and thromboelastography (TEG), have been evaluated as predictive markers of hypercoagulability in MPNs. However, validation of these and other laboratory surrogates is lacking. The objective of this systematic review is to summarize the literature and to examine the utility of global hemostatic assays as predictive markers for hypercoagulability and occurrence of thrombosis in MPNs.

Methods: A systematic literature search was performed using MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews from inception to April 14, 2022, restricted to the English language. Criteria for eligible studies were studies that reported the results of TGA, ROTEM, and TEG in ET, PV, or PMF adult patients. Studies were required to have ≥10 subjects in each group. Two reviewers (A.T. and T.C.) independently searched the literature and extracted data from eligible studies. Disagreements were resolved by consensus or a third reviewer (C.K) when necessary. The methodological quality of included studies was appraised using the Newcastle - Ottawa Quality Assessment Scale (NOS) assessment tool. A meta-analysis was planned given sufficient data from included studies. The study protocol is available on PROSPERO (CRD42022325893).

Results: A total of 815 records were retrieved from the literature search. After screening by title and abstract, 745 records were excluded. The remaining 70 references underwent full text review, 30 of which met eligibility criteria and were included in the analysis. These 30 studies (15 full-texts and 15 abstracts) collectively enrolled 926 controls, 700 ET, 556 PV, and 192 PMF patients. The number of studies that reported TGA, ROTEM, and TEG were 26, 3, and 3, respectively. The number of participants in each study ranged from 25 to 212 including the controls. Two studies originated from Oceania, 1 from North America, and 27 from Europe. JAK2 mutation was reported only in 15 studies and only two studies reported quantitative variant allele frequencies. There were 5 studies that did not have healthy controls. All of these studies were cross sectional analyses; no longitudinal prospective age/sex matched controlled studies have been published.

Of the 26 TGA studies, comparison of TGA parameters were made between MPN vs. controls (15 studies), MPN with thrombosis vs. without thrombosis (3 studies), JAK2 positive vs. negative MPNs (11 studies), and MPN with vs. without treatment (6 studies). There is heterogeneity in the types of samples (platelet poor plasma or platelet rich plasma), trigger reagents (amounts of tissue factor, phospholipid, and thrombomodulin), and the automated platforms that were used.

Significant differences in the endogenous thrombin potential (9 studies), peak height (10 studies), lag time (6 studies), and velocity index (3 studies) were found between MPN and controls.

Thirteen studies reported rates of arterial and venous thrombotic outcomes in MPN patients, which ranged from 19 to 45%. Only 1 study found an association of TGA parameters and thrombosis. Two ROTEM studies showed ET having higher maximum clot firmness and shorter clot formation time compared to controls. Two TEG studies showed increased alpha angle between MPN and controls and between PV and controls.

Risk of bias assessment with NOS showed summary scores ranging 1-3 for abstracts and 4-9 for full text studies.

We were not able to perform meta-analysis due to insufficient data and the high heterogeneity in study objectives and laboratory methods.

Conclusions: Studies were limited by small sample sizes and inconsistent results and do not comprehensively reflect the effects of treatment, duration of disease, or clonal status. The role of these global hemostatic assays in predicting thrombotic events remains inconclusive. Future studies are necessary to establish the usefulness and validity of these surrogate markers as predictors of thrombotic complications in the MPNs.

No relevant conflicts of interest to declare.