New diagnostic tests are needed for the early detection of acute stroke. Elevated plasma free fatty acids (FFA) are an early marker of stroke, but current methods to measure FFA are too slow to be used for stroke detection. In contrast, measurements of the unbound fraction of plasma FFA (FFAu) can be completed in seconds using FFAu-specific fluorescent probes. In the proposed Phase I study, a rapid test for stroke detection will be developed using probes sensitive to changes in the plasma FFAu profile. This work will extend preliminary investigations that demonstrated significant discrimination between stroke patients and mimics using FFAu probes made from fluorescently labeled mutants of rat intestinal fatty acid binding protein.
The specific aims of the proposed study are to (1) develop new probes that will enhance FFAu profiling accuracy and improve diagnostic performance and (2) validate the performance of the multiple-probe test for stroke with samples from the Specialized Programs of Translational Research in Acute Stroke (SPOTRIAS) repository.
In Aim 1, a new set of probes specific for arachidonate, linolenate, docosahexaenoate, and myristate will be characterized and tested for profiling accuracy use model FFAu mixtures. These probes will be used with a previous probe set with specificities for stearate, oleate, linoleate, palmitate, and palmitoleate to determine the plasma FFAu profiles in 33 stroke patients and 22 mimics. Using linear discriminant analysis, stroke classifiers will be constructed from the FFAu profiles and from the set of probe responses. High-throughput screening of libraries of mutant probes will be used to develop additional probes that will enable further improvements in profiling accuracy.
In Aim 2, the classifiers developed in Aim 1 will be validated with a new set of plasma samples from 65 stroke patients and 45 mimics from the SPOTRIAS repository. Samples will be measured and classified as stroke or mimic without knowledge of the patient diagnoses. Receiver operating characteristic curves will be used to assess diagnostic performance after the patient diagnoses are revealed. Finally, a new classifier will be generated from the pooled probe data for all 165 samples. This new classifier will be tested in Phase II with a larger sample set.
The early diagnosis of stroke is critical for effective therapeutic intervention, yet no rapid test for stroke is available. To meet this clinical need, a rapid and accurate test that determines the profile of fatty acids in blood will be developed for the early detection of ischemic and hemorrhagic stroke. The use of this test would potentially reduce neurological damage from ischemic stroke by reducing the time required to diagnose and treat stroke patients.