Cancer Prediction With Machine Learning of Thrombi From Thrombectomy in Stroke: Multicenter Development and Validation.

BACKGROUND: We aimed to develop and validate machine learning models to diagnose patients with ischemic stroke with cancer through the analysis of histopathologic images of thrombi obtained during endovascular thrombectomy. METHODS: This was a retrospective study using a prospective multicenter registry which enrolled consecutive patients with acute ischemic stroke from South Korea who underwent endovascular thrombectomy. This study included patients admitted between July 1, 2017 and December 31, 2021 from 6 academic university hospitals. Whole-slide scanning was performed for immunohistochemically stained thrombi. Machine learning models were developed using transfer learning with image slices as input to classify patients into 2 groups: cancer group or other determined cause group. The models were developed and internally validated using thrombi from patients of the primary center, and external validation was conducted in 5 centers. The model was also applied to patients with hidden cancer who were diagnosed with cancer within 1 month of their index stroke. RESULTS: The study included 70 561 images from 182 patients in both internal and external datasets (119 patients in internal and 63 in external). Machine learning models were developed for each immunohistochemical staining using antibodies against platelets, fibrin, and erythrocytes. The platelet model demonstrated consistently high accuracy in classifying patients with cancer, with area under the receiver operating characteristic curve of 0.986 (95% CI, 0.983-0.989) during training, 0.954 (95% CI, 0.937-0.972) during internal validation, and 0.949 (95% CI, 0.891-1.000) during external validation. When applied to patients with occult cancer, the model accurately predicted the presence of cancer with high probabilities ranging from 88.5% to 99.2%. CONCLUSIONS: Machine learning models may be used for prediction of cancer as the underlying cause or detection of occult cancer, using platelet-stained immunohistochemical slide images of thrombi obtained during endovascular thrombectomy.

Characterization of Ferric Chloride-Induced Arterial Thrombosis Model of Mice and the Role of Red Blood Cells in Thrombosis Acceleration.

PURPOSE: The ferric chloride (FeCl3)-induced thrombosis model is widely used for thrombosis research. However, it lacks standardization with uncertainty in the exact mechanism of thrombosis. This study aimed to characterize thrombus formation in a mouse model. MATERIALS AND METHODS: We investigated thrombus formation and stability using various FeCl3 concentrations (10%, 20%, 30%, 40%, and 50%, w/v) in carotid arteries of the Institute of Cancer Research (ICR) and C57BL/6N mice using the FeCl3-induced thrombosis model. We also investigated thrombus histopathology using immunohistochemistry and electron microscopy. RESULTS: Higher FeCl3 concentrations induced dose-dependent, faster, larger, and more stable thrombus formation in both strains of mice. However, the ICR mice showed better dose-responses in thrombus formation and stability compared to the C57BL/6N mice. Thrombi were fibrin- and platelet-rich without significant changes across FeCl3 concentrations. However, the content of red blood cells (RBCs) increased with increasing FeCl3 concentrations (p for trend <0.001) and inversely correlated with time to occlusion (r=-0.65, p<0.001). While platelets and fibrin were evenly distributed over the thrombus, RBCs were predominantly located near the FeCl3 treatment area. Transmission electron microscopy showed that RBCs attached to and were surrounded by aggregates of degranulated platelets, suggesting their potential role in platelet activation. CONCLUSION: Faster and larger thrombus formation is induced in a dose-dependent manner by a wide range of FeCl3 concentrations, but the stable thrombus formation requires higher FeCl3 concentrations. Mouse strain affects thrombus formation and stability. RBCs and their interaction with platelets play a key role in the acceleration of FeCl3-induced thrombosis.

Effects of exercise training and TrkB blockade on cardiac function and BDNF-TrkB signaling postmyocardial infarction in rats.

Exercise training is beneficial for preserving cardiac function postmyocardial infarction (post-MI), but the underlying mechanisms are not well understood. We investigated one possible mechanism, brain-derived neurotrophic factor (BDNF)-tropomyosin-related kinase B (TrkB) signaling, with the TrkB blocker ANA-12 (0.5 mg·kg-1·day-1). Male Wistar rats underwent sham surgery or ligation of the left descending coronary artery. The surviving MI rats were allocated as follows: sedentary MI rats treated with vehicle, exercise-trained MI rats treated with vehicle, and exercise-trained MI rats treated with ANA-12. Exercise training was done 5 days/wk for 4 wk on a motor-driven treadmill. At the end, left ventricular (LV) function was evaluated by echocardiography and a Millar catheter. Mature BDNF and downstream effectors of BDNF-TrkB signaling, Ca2+/calmodulin-dependent protein kinase II (CaMKII), Akt, and AMP-activated protein kinase (AMPK), were assessed in the noninfarct area of the LV by Western blot analysis. Exercise training increased stroke volume and cardiac index and attenuated the decrease in ejection fraction (EF) and increase in LV end-diastolic pressure post-MI. ANA-12 blocked the improvement of EF and attenuated the increases in stroke volume and cardiac index but did not affect LV end-diastolic pressure. Exercise training post-MI prevented decreases in mature BDNF, phosphorylated (p-)CaMKII, p-Akt, and p-AMPKα expression. These effects were all blocked by ANA-12 except for p-AMPKα. In conclusion, the exercise-induced improvement of EF is mediated by the BDNF-TrkB axis and the downstream effectors CaMKII and Akt. BDNF-TrkB signaling appears to contribute to the improvement in systolic function by exercise training. NEW & NOTEWORTHY Exercise training improves ejection fraction and left ventricular end-diastolic pressure (LVEDP) and increases stroke volume and cardiac index in rats postmyocardial infarction (post-MI). The improvement of EF but not LVEDP is mediated by activation of the brain-derived neurotrophic factor (BDNF)-tropomyosin-related kinase B (TrkB) axis and downstream effectors Ca2+/calmodulin-dependent protein kinase II (CaMKII) and Akt. This suggests that activation of BDNF-TrkB signaling and CaMKII and Akt is a promising target to attenuate progressive cardiac dysfunction post-MI.

Effects of exercise training on brain-derived neurotrophic factor in skeletal muscle and heart of rats post myocardial infarction.

NEW FINDINGS: What is the central question of this study? Exercise training increases brain-derived neurotrophic factor (BDNF) in the hippocampus, which depends on a myokine, fibronectin type III domain-containing protein 5 (FNDC5). Whether exercise training after myocardial infarction induces parallel increases in FNDC5 and BDNF expression in skeletal muscle and the heart has not yet been studied. What is the main finding and its importance? Exercise training after myocardial infarction increases BDNF protein in skeletal muscle and the non-infarct area of the LV without changes in FNDC5 protein, suggesting that BDNF is not regulated by FNDC5 in skeletal muscle and heart. An increase in cardiac BDNF may contribute to the improvement of cardiac function by exercise training. Exercise training after myocardial infarction (MI) attenuates progressive left ventricular (LV) remodelling and dysfunction, but the peripheral stimuli induced by exercise that trigger these beneficial effects are still unclear. We investigated as possible mediators fibronectin type III domain-containing protein 5 (FNDC5) and brain-derived neurotrophic factor (BDNF) in the skeletal muscle and heart. Male Wistar rats underwent either sham surgery or ligation of the left descending coronary artery, and surviving MI rats were allocated to either a sedentary (Sed-MI) or an exercise group (ExT-MI). Exercise training was done for 4 weeks on a motor-driven treadmill. At the end, LV function was evaluated, and FNDC5 and BDNF mRNA and protein were assessed in soleus muscle, quadriceps and non-, peri- and infarct areas of the LV. At 5 weeks post MI, FNDC5 mRNA was decreased in soleus muscle and all areas of the LV, but FNDC5 protein was increased in the soleus muscle and the infarct area. Mature BDNF (mBDNF) protein was decreased in the infarct area without a change in mRNA. Exercise training attenuated the decrease in ejection fraction and the increase in LV end-diastolic pressure post MI. Exercise training had no effect on FNDC5 mRNA and protein, but increased mBDNF protein in soleus muscle, quadriceps and the non-infarct area of the LV. The mBDNF protein in the non-infarct area correlated positively with ejection fraction and inversely with LV end-diastolic pressure. In conclusion, mBDNF is induced by exercise training in skeletal muscle and the non-infarct area of the LV, which may contribute to improvement of muscle dysfunction and cardiac function post MI.