Background: Immune Thrombocytopenia Purpura (ITP) is defined by a low platelet count in the absence of any known causes and affects over 4,000 US children and 8,000 adults each year (Terrell et al, 2010). Deciding when and how to treat these patients remains difficult as there is no biomarker or diagnostic test that identifies which patients will self-resolve and which are at risk for major bleeding and life-threatening intracranial hemorrhage (~10%). In addition, the medications used to treat ITP all have significant side effects, forcing clinicians to balance risks associated with bleeding and medication. As such, an ongoing debate in the field of clinical hematology centers around which patients require therapy (Flores & Buchanan, 2013)(Cooper 2017)(Gralnek 2008). Here, we show that a new quantitative measurement of platelet function, which is independent of known platelet biomarkers (Myers et al, 2017), has diagnostic potential in identifying bleeding risk in ITP patients. Specifically, using high-throughput platelet contraction cytometry (PCC) measurements of individual platelet forces (Myers et al, 2017), we found that low platelet forces strongly correlate with bleeding symptoms. Unlike existing tests of platelet function that use bulk measurements, our test operates at the single cell level and therefore does not depend on the platelet count, enabling a direct comparison of individual platelet function in health versus disease states. In the broad context of overall function, our findings of impaired force generation agree with previous research demonstrating that impaired platelet function, and not low platelet count, correlates with bleeding in ITP patients (Frelinger et al, 2015). In addition, as the first demonstration of a correlation between single cell force and disease, our novel approach may represent an entirely new class of "physics-based" diagnostics.

Platelet Contraction Cytometry (PCC): Within the PCC, a single platelet attaches, spreads, and applies contractile force to a pair of fibrinogen microdots that are attached to a moveable, spring-like, surface. Since the applied platelet contractile force is directly proportional to the microdot displacement, the force is calculated from a single fluorescence image of the platelet. Using microfabrication technology, thousands of microdots are created on a single device to enable high-throughput measurements in tightly controlled mechanical, biochemical, and shear microenvironments (Fig 1). Here, gel-filtered platelets from patients are activated (1U/mL thrombin), plated on the device with a moderately stiff surface (75 kPa), and measured as described previously (Myers et al, 2017).

Results: In a cohort of ITP patients (n = 27), we observed that patients with bleeding and/or bruising symptoms exhibited significantly lower average platelet contraction forces than asymptomatic patients regardless of platelet count. Using an average force cutoff value of 26nN, we found that low forces identified bleeding in ITP with 100% sensitivity and 89.4% specificity. From a mechanistic perspective, our preliminary data also suggests that mean platelet volume does not correlate with platelet force or function, although further studies on a single platelet level are needed to confirm this result. While this data indicates that patients with bleeding symptoms have platelets with low forces, it is unclear whether low forces correlate with bleeding or whether the patient always has had low platelet forces that therefore render them susceptible to bleeding. Our preliminary prospective data suggests that low platelet forces correlate with the bleeding symptoms themselves and return to higher values when bleeding symptoms cease.

Importantly, as our approach enables "single-cell" examinations of platelet forces, we identified different platelet subpopulations unique to ITP. Healthy individuals have a distribution of contractile forces with a single peak, while ITP patients tend to have two prominent peaks, one with a lower contractile force and one with a higher one. Surprisingly, patients with bleeding symptoms only have a low force peak, suggesting that the lack of highly contractile platelets may be a biophysical biomarker for bleeding (Fig 2). Ongoing studies are focused on using this finding to further improve specificity as well as elucidating the mechanistic underpinnings of low platelet contraction in ITP.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution