Karl Sjölin: Finding stroke with a blood test
- Date: 4 October 2024, 09:00
- Location: H:son-Holmdahlsalen, Akademiska Sjukhuset, Ing 100, 2 tr, Uppsala
- Type: Thesis defence
- Thesis author: Karl Sjölin
- External reviewer: Katarina Jood
- Supervisors: Joachim Burman, Kim Kultima, Anders Larsson, Signild Åsberg
- Research subject: Neurology
- DiVA
Abstract
In contrast to many other diseases and conditions, there is no established blood-based biomarker to aid in the diagnosis, prognosis, or outcome prediction of stroke. The neurospecific proteins glial fibrillary acidic protein (GFAP), myelin basic protein, neurofilament light (NFL), tau, and ubiquitin carboxy-terminal hydrolase L1 are released into blood in response to injurious processes affecting the central nervous system. This thesis aims to enhance the understanding of if, how, and when these biomarkers can provide important information in stroke and stroke-related disorders and which should be the focus for further translation into a useful blood test for stroke.
Firstly, we determined how and at which concentrations these biomarkers are distributed in the human CNS. We found substantial variation between brain regions, indicating that these biomarkers' circulating levels are likely affected by both the size and location of a cerebral insult.
After that, we investigated how plasma levels of these biomarkers change during the first week after an ischemic stroke and determined the optimal time point for assessing infarct volume. Undoubtedly, GFAP was the most optimal biomarker to assess infarct volume in the acute phase, while NFL was better suited to evaluate infarct volume one week to three months after symptom onset.
The findings indicated that NFL holds information long after a cerebrovascular event, so we analyzed plasma NFL in patients undergoing the cardiac procedure transcatheter aortic valve implantation, which is associated with a high frequency of relatively small and covert brain infarcts. We found that NFL increased by 60% after the procedure, which in approximative numbers corresponds to 1 cm3 infarcted brain tissue, similar to the previously reported mean lesion size after the procedure, indicating that NFL may contribute to the detection of procedure-related insults.
Finally, we analyzed serum NFL in patients with atrial fibrillation (AF), a cardiac disease associated with both overt and covert brain infarcts, and in matched controls. We discovered that patients with AF had slightly elevated levels of NFL and that patients with ongoing AF rhythm had the highest levels, indicating that the cerebrovascular pathologies associated with AF may, at least in part, be reflected by NFL.
In summary, this thesis has contributed to the understanding of how and when these biomarkers provide information about stroke and stroke-related disorders. Future studies should aim to further NFL and GFAP into the clinical management of cerebrovascular disease.