ICSDA: a multi-modal deep learning model to predict breast cancer recurrence and metastasis risk by integrating pathological, clinical and gene expression data.

Breast cancer patients often have recurrence and metastasis after surgery. Predicting the risk of recurrence and metastasis for a breast cancer patient is essential for the development of precision treatment. In this study, we proposed a novel multi-modal deep learning prediction model by integrating hematoxylin & eosin (H&E)-stained histopathological images, clinical information and gene expression data. Specifically, we segmented tumor regions in H&E into image blocks (256 × 256 pixels) and encoded each image block into a 1D feature vector using a deep neural network. Then, the attention module scored each area of the H&E-stained images and combined image features with clinical and gene expression data to predict the risk of recurrence and metastasis for each patient. To test the model, we downloaded all 196 breast cancer samples from the Cancer Genome Atlas with clinical, gene expression and H&E information simultaneously available. The samples were then divided into the training and testing sets with a ratio of 7: 3, in which the distributions of the samples were kept between the two datasets by hierarchical sampling. The multi-modal model achieved an area-under-the-curve value of 0.75 on the testing set better than those based solely on H&E image, sequencing data and clinical data, respectively. This study might have clinical significance in identifying high-risk breast cancer patients, who may benefit from postoperative adjuvant treatment.

Electroacupuncture Pretreatment Prevents Cognitive Impairment Induced by Cerebral Ischemia-Reperfusion via Adenosine A1 Receptors in Rats.

A previous study has demonstrated that pretreatment with electroacupuncture (EA) induces rapid tolerance to focal cerebral ischemia. In the present study, we investigated whether adenosine receptor 1 (A1 R) is involved in EA pretreatment-induced cognitive impairment after focal cerebral ischemia in rats. Two hours after EA pretreatment, focal cerebral ischemia was induced by middle cerebral artery occlusion for 120 min in male Sprague-Dawley rats. The neurobehavioral score, cognitive function [as determined by the Morris water maze (MWM) test], neuronal number, and the Bax/Bcl-2 ratio was evaluated at 24 h after reperfusion in the presence or absence of CCPA (a selective A1 receptor agonist), DPCPX (a selective A1 receptor antagonist) into left lateral ventricle, or A1 short interfering RNA into the hippocampus area. The expression of the A1 receptor in the hippocampus was also investigated. The result showed that EA pretreatment upregulated the neuronal expression of the A1 receptor in the rat hippocampus at 90 min. And EA pretreatment reversed cognitive impairment, improved neurological outcome, and inhibited apoptosis at 24 h after reperfusion. Pretreatment with CCPA could imitate the beneficial effects of EA pretreatment. But the EA pretreatment effects were abolished by DPCPX. Furthermore, A1 receptor protein was reduced by A1 short interfering RNA which attenuated EA pretreatment-induced cognitive impairment.

[Study on efficacy specificity of electroacupuncture at "Zusanli "(ST36) and "Baihui" (GV20) in cerebral ischemia and inflammatory pain rats].

OBJECTIVE: To explore the efficacy difference between electroacupuncture (EA) at "Zusanli" (ST36) and "Baihui" (GV20) for inflammatory pain and cerebral ischemia-reperfusion injury (CIRI) in rats. METHODS: In 1st part of this study, 90 male SD rats were randomly divided into sham-operation, model (induced by occlusion of the middle cerebral artery and reperfusion), GV20 EA, ST36 EA，and sham EA groups (n=16 in each group). In the 2nd part of the study, 40 male SD rats were randomized into saline injection (control), inflammatory pain model (subcutaneous injection of complete Freund's adjuvant ［CFA］ into the right paw), ST36 EA, GV20 EA, and sham EA groups (n=8 in each group). In these two parts, EA (2 Hz/15 Hz, 1 mA) was applied to ST36 or GV20. The mechanical withdrawal threshold (MWT) and thermal withdrawal latency （TWL） were detected 2.5 h after administration of CFA by using Von Frey and plantar tester, respectively. The neurological deficit scores (NDS) were assessed by using Longa's method and the infarct size of the brain assessed after staining with 2% triphenyltetrazolium chloride (TTC). The expression of c-fos protein in the dorsal horns (DHs) of the spinal cord was detected by immunohistochemistry. RESULTS: (1) Twenty-four hours following CIRI, the NDS and infarct volume were significantly increased in the model group compared with the sham-operation group (P<0.01), and obviously decreased in the GV20 EA and ST36 EA groups relevant to the CIRI model group (P<0.05, P<0.01). There were no significant differences between the two EA groups in the NDS and infarct volume levels (P>0.05). (2) After administration of CFA, both the MPT and TPT were notably decreased in the inflammatory pain model group in contrast to the saline-injection group (P<0.01), but were considerably increased in both ST36 EA and GV20 EA groups (P<0.05), rather than in the sham EA group (P>0.05). The number of c-fos positive cells was significantly increased in the medial half of I-II and III-IV lamina of DHs in the L4-L6 segments of spinal cord in the inflammatory pain model group relevant to the saline-injection group (P<0.01,P<0.05), and was remarkably decreased in the lamina I-II (not in the deeper lamina) in both ST36 EA and GV20 EA groups (P<0.01), rather than in the sham EA group (P>0.05). No significant differences were found in the number of c-fos positive cells between the ST36 EA and GV20 EA groups (P>0.05). CONCLUSION: Our data do not support the specificity of functions at least between GV20 EA and ST36 EA in both CIRI and inflammatory pain model rats. This is the first study reporting the effect of EA at GV20 for relieving CFA-induced inflammatory pain.

Islet-cell autoantigen 69 mediates the antihyperalgesic effects of electroacupuncture on inflammatory pain by regulating spinal glutamate receptor subunit 2 phosphorylation through protein interacting with C-kinase 1 in mice.

Electroacupuncture (EA) is widely used in clinical settings to reduce inflammatory pain. Islet-cell autoantigen 69 (ICA69) has been reported to regulate long-lasting hyperalgesia in mice. ICA69 knockout led to reduced protein interacting with C-kinase 1 (PICK1) expression and increased glutamate receptor subunit 2 (GluR2) phosphorylation at Ser880 in spinal dorsal horn. In this study, we evaluated the role of ICA69 in the antihyperalgesic effects of EA and the underlying mechanism through regulation of GluR2 and PICK1 in spinal dorsal horn. Hyperalgesia was induced in mice with subcutaneous plantar injection of complete Freund adjuvant (CFA) to cause inflammatory pain. Electroacupuncture was then applied for 30 minutes every other day after CFA injection. When compared with CFA group, paw withdrawal frequency of CFA+EA group was significantly decreased. Remarkable increases in Ica1 mRNA expression and ICA69 protein levels on the ipsilateral side were detected in the CFA+EA group. ICA69 expression reached the peak value around day 3. More importantly, ICA69 deletion impaired the antihyperalgesic effects of EA on GluR2-p, but PICK1 deletion could not. Injecting ICA69 peptide into the intrathecal space of ICA69-knockout mice mimicked the effects of EA analgesic and inhibited GluR2-p. Electroacupuncture had no effects on the total protein of PICK1 and GluR2. And, EA could increase the formation of ICA69-PICK1 complexes and decrease the amount of PICK1-GluR2 complexes. Our findings indicate that ICA69 mediates the antihyperalgesic effects of EA on CFA-induced inflammatory pain by regulating spinal GluR2 through PICK1 in mice.