A longitudinal evaluation of oxidative stress - mitochondrial dysfunction - ferroptosis genes in anthracycline-induced cardiotoxicity.

BACKGROUND: Antineoplastic medications, including doxorubicin, idarubicin, and epirubicin, have been found to adversely affect the heart due to oxidative stress - mitochondrial dysfunction - ferroptosis (ORMFs), which act as contributing attributes to anthracycline-induced cardiotoxicity. To better understand this phenomenon, the time-resolved measurements of ORMFS genes were analyzed in this study. METHODS: The effect of three anthracycline drugs on ORMFs genes was studied using a human 3D cardiac microtissue cell model. Transcriptome data was collected over 14 days at two doses (therapeutic and toxic). WGCNA identified key module-related genes, and functional enrichment analysis investigated the biological processes quantified by ssGSEA, such as immune cell infiltration and angiogenesis. Biopsies were collected from heart failure patients and control subjects. GSE59672 and GSE2965 were collected for validation. Molecular docking was used to identify anthracyclines's interaction with key genes. RESULTS: The ORMFs genes were screened in vivo or in vitro. Using WGCNA, six co-expressed gene modules were grouped, with MEblue emerging as the most significant module. Eight key genes intersecting the blue module with the dynamic response genes were obtained: CD36, CDH5, CHI3L1, HBA2, HSD11B1, OGN, RPL8, and VWF. Compared with control samples, all key genes except RPL8 were down-regulated in vitro ANT treatment settings, and their expression levels varied over time. According to functional analyses, the key module-related genes were engaged in angiogenesis and the immune system pathways. In all ANT-treated settings, ssGSEA demonstrated a significant down-regulation of angiogenesis score and immune cell activity, including Activated CD4 T cell, Immature B cell, Memory B cell, Natural killer cell, Type 1 T helper cell, and Type 2 T helper cell. Molecular docking revealed that RPL8 and CHI3L1 show significant binding affinity for anthracyclines. CONCLUSION: This study focuses on the dynamic characteristics of ORMFs genes in both human cardiac microtissues and cardiac biopsies from ANT-treated patients. It has been highlighted that ORMFs genes may contribute to immune infiltration and angiogenesis in cases of anthracycline-induced cardiotoxicity. A thorough understanding of these genes could potentially lead to improved diagnosis and treatment of the disease.

Drug-induced musical hallucination.

Musical hallucination is a rare perceptual phenomenon wherein individuals hear music in the absence of external auditory stimuli. This phenomenon occurs across diverse medical conditions and can be triggered by some drugs. The underlying mechanism of drug-induced hallucination is unknown. This study explores drug-induced musical hallucination through a literature review, aiming to investigate its pathophysiology and potential treatment modalities. A literature search was conducted until January 2024 using databases PubMed, WorldCat, Google Scholar, and DOAJ, with keywords "drugs induced musical hallucination" or "drugs" combined with "musical hallucination." The search yielded 24 articles which met inclusion criteria, encompassing 27 cases. The average patient age was 58.3 years, with 67.9% females. Prevalent conditions among cases included hearing impairments, psychiatric disorders, cancers, and neurodegenerative conditions. Common trigger drugs comprised antidepressants, opioids, anti-Parkinson drugs, ketamine, and voriconazole. Musical hallucination descriptions varied widely, and 6 patients reported concurrent visual hallucinations. The onset of symptoms ranged from 75 min to 240 days. Treatment strategies included termination of trigger drugs, dosage reduction, alteration of administration routes or formula, switching to similar drugs, or addition of antidepressants, sedatives, or atypical antipsychotic medications. Musical hallucinations completely disappeared in 24/27 (88.9%) patients but continued in 3/27 (11.1%) patients. The current study concludes that drug-induced musical hallucination may arise from altering neurotransmitter/receptor balance and intricate interactions between trigger drugs and underlying conditions.

HMMF: a hybrid multi-modal fusion framework for predicting drug side effect frequencies.

BACKGROUND: The identification of drug side effects plays a critical role in drug repositioning and drug screening. While clinical experiments yield accurate and reliable information about drug-related side effects, they are costly and time-consuming. Computational models have emerged as a promising alternative to predict the frequency of drug-side effects. However, earlier research has primarily centered on extracting and utilizing representations of drugs, like molecular structure or interaction graphs, often neglecting the inherent biomedical semantics of drugs and side effects. RESULTS: To address the previously mentioned issue, we introduce a hybrid multi-modal fusion framework (HMMF) for predicting drug side effect frequencies. Considering the wealth of biological and chemical semantic information related to drugs and side effects, incorporating multi-modal information offers additional, complementary semantics. HMMF utilizes various encoders to understand molecular structures, biomedical textual representations, and attribute similarities of both drugs and side effects. It then models drug-side effect interactions using both coarse and fine-grained fusion strategies, effectively integrating these multi-modal features. CONCLUSIONS: HMMF exhibits the ability to successfully detect previously unrecognized potential side effects, demonstrating superior performance over existing state-of-the-art methods across various evaluation metrics, including root mean squared error and area under receiver operating characteristic curve, and shows remarkable performance in cold-start scenarios.

Characteristics of Ocular Inflammatory Side Effects Associated with Immune Checkpoint Inhibitors in a Northern California Population.

PURPOSE: To characterize the ocular inflammatory side effects associated with immune checkpoint inhibitor (CPI) treatment in a Northern California population. DESIGN: Retrospective case series. PARTICIPANTS: Patients receiving CPI within an integrated healthcare delivery system. METHODS: All patients within Kaiser Permanente Northern California receiving CPI between January 1, 2012 and November 1, 2018 were identified. Medical records of those seen in the ophthalmology clinic at least once were retrospectively reviewed. MAIN OUTCOME MEASURES: Type and duration of ocular inflammation, indication for and exposure to CPI, time from exposure to diagnosis of ocular inflammation. RESULTS: 31 cases of ocular inflammation were identified in 5061 patients (0.61%) receiving CPI. Mean ± SD age was 67 ± 11.9 (range 38-89). Mean time from exposure to diagnosis was 6.8 ± 5.5 months (range 0.5-17). 87% of cases were bilateral, and 43% of cases were chronic. Average ophthalmology follow-up was 16 ± 18 months (range 0-71). 16/31 (52%) had anterior uveitis, 7/31 (23%) had serous retinal detachment or panuveitis resembling Vogt-Koyanagi-Harada syndrome, 4/31 (13%) had papillitis, and 6/31 (19%) had diplopia or ocular motility defect. There was one case each (3.2%) of melanoma associated retinopathy, corneal edema, granulomatous lacrimal gland enlargement, and choroidal neovascularization. CONCLUSIONS: Ocular inflammation is a rare immune associated side effect of CPI treatment, the most common manifestation of which is anterior uveitis.

MSDSE: Predicting drug-side effects based on multi-scale features and deep multi-structure neural network.

Unexpected side effects may accompany the research stage and post-marketing of drugs. These accidents lead to drug development failure and even endanger patients' health. Thus, it is essential to recognize the unknown drug-side effects. Most existing methods in silico find the answer from the association network or similarity network of drugs while ignoring the drug-intrinsic attributes. The limitation is that they can only handle drugs in the maturation stage. To be suitable for early drug-side effect screening, we conceive a multi-structural deep learning framework, MSDSE, which synthetically considers the multi-scale features derived from the drug. MSDSE can jointly learn SMILES sequence-based word embedding, substructure-based molecular fingerprint, and chemical structure-based graph embedding. In the preprocessing stage of MSDSE, we project all features to the abstract space with the same dimension. MSDSE builds a bi-level channel strategy, including a convolutional neural network module with an Inception structure and a multi-head Self-Attention module, to learn and integrate multi-modal features from local to global perspectives. Finally, MSDSE regards the prediction of drug-side effects as pair-wise learning and outputs the pair-wise probability of drug-side effects through the inner product operation. MSDSE is evaluated and analyzed on benchmark datasets and performs optimally compared to other baseline models. We also set up the ablation study to explain the rationality of the feature approach and model structure. Moreover, we select model partial prediction results for the case study to reveal actual capability. The original data are available at http://github.com/yuliyi/MSDSE.

Skin Rash as a Side Effect of Bortezomib: A Case Report.

Bortezomib is a novel proteasome inhibitor widely used in the treatment of multiple myeloma, especially in the case of recurrent, relapsing, or refractory myeloma. Its common side effects include nausea, vomiting, neuropathic pain, and hemorrhage. Cutaneous manifestations are considered rare. Here we report a case of a 72-year-old female patient, and a known case of multiple myeloma on VRD protocol (bortezomib [Velcade] combined with lenalidominde [Revlimid] and dexamethasone) after relapse. The patient presented with a complaint of raised, itchy, erythematous skin rash, starting in the soles and palms and then spreading to the whole body. A skin biopsy confirmed that the lesions were due to an allergic reaction. The patient was admitted as a case of sepsis and died. Skin rash is considered a rare side effect of bortezomib, with variable presentation and onset. The mainstay of treatment is corticosteroids.

Nocturnal Leg Cramping Caused by Carnitine Deficiency Due to Long-Term Pivalate Antibiotics Administration in a Patient With Chronic Kidney Disease.

Carnitine deficiency is a known cause of leg cramps and is sometimes observed in patients taking certain medications such as pivalate-containing antibiotics. A 69-year-old Japanese woman presented with a progression of painful involuntary nocturnal leg cramping. She had been taking cefcapene-pivoxil for six months. Serum-free carnitine (FC) and acylcarnitine levels were decreased. Then, carnitine deficiency due to long-term pivalate-containing antibiotics administration was diagnosed. After initiating oral L-carnitine treatment, her symptoms improved. It should be aware of carnitine deficiency if a patient taking pivalate-containing antibiotics presents with leg cramping.

Efficacy of Bee Products (Anzer Honey, Pollen and Propolis) in Detection and Healing of Damage Induced by Antidiabetic Drug Vildagliptin/Metformin Hydrochloride in Healthy Human Pancreatic Cells: Cytotoxic, Genotoxic and Biochemical Studies.

BACKGROUND AND OBJECTIVE: Although drugs are powerful therapeutic agents, they have a range of side effects. These side effects are sometimes cellular and not clinically noticeable. Vildagliptin/metformin hydrochloride is one of the most widely used oral antidiabetic drugs with two active ingredients. In this study, we investigated its harmful effects on the metabolic activation system in healthy human pancreatic cells "hTERT-HPNE", and we aimed to improve these harmful effects by natural products. To benefit from the healing effect, we used the unique natural products produced by the bees of the Anzer Plateau in the Eastern Black Sea Region of Turkey. METHODS: Cytotoxic and genotoxic effects of the drug were investigated by different tests, such as MTT, flow cytometry-apoptosis and comet assays. Anzer honey, pollen and propolis were analyzed by gas chromatography/mass spectrometry (G/C-MS). A total of 19 compounds were detected, constituting 99.9% of the samples. RESULTS: The decrease in cell viability at all drug concentrations was statistically significant compared to the negative control (P<0.05). A statistically significant decrease was detected in the apoptosis caused by vildagliptin/metformin hydrochloride with the supplementation of Anzer honey, pollen and propolis in hTERT-HPNE cells (P<0.05). CONCLUSION: This study can contribute to other studies testing the healing properties of natural products against the side effects of oral antidiabetics in human cells. In particular, Anzer honey, pollen and propolis can be used as additional foods to maintain cell viability and improve heal damage and can be evaluated against side effects in other drug studies.

Identification of drug-side effect association via correntropy-loss based matrix factorization with neural tangent kernel.

Adverse drug reactions include side effects, allergic reactions, and secondary infections. Severe adverse reactions can cause cancer, deformity, or mutation. The monitoring of drug side effects is an important support for post marketing safety supervision of drugs, and an important basis for revising drug instructions. Its purpose is to timely detect and control drug safety risks. Traditional methods are time-consuming. To accelerate the discovery of side effects, we propose a machine learning based method, called correntropy-loss based matrix factorization with neural tangent kernel (CLMF-NTK), to solve the prediction of drug side effects. Our method and other computational methods are tested on three benchmark datasets, and the results show that our method achieves the best predictive performance.

Nivolumab-induced Lambert-Eaton myasthenic syndrome: an immune-related adverse event in nonsmall cell lung cancer.

Nivolumab is an anti-programmed cell death protein 1 (anti-PD-1) monoclonal antibody and was the first immune checkpoint inhibitor drug approved for use in advanced non-small cell lung cancer (NSCLC). In this report, we describe a rare case of Lambert-Eaton myasthenic syndrome (LEMS), which developed as a side effect of nivolumab in a patient with metastatic lung squamous cell carcinoma. Our patient, who was previously treated with nivolumab for metastatic squamous cell carcinoma of the lung, appeared with a headache, swollen face, dysarthria, asthenia, xerostomia, and drooping eyelid. Early testing indicated no thymomas or newly developing tumors in whole-body scans, and the blood workup was normal. We came to the conclusion that nivolumab-induced LEMS was the cause of the symptoms after performing nerve conduction investigations ruling out other differentials. We believe our clinical experience of this rare and unexpected adverse event should be shared.

CU06-1004 as a promising strategy to improve anti-cancer drug efficacy by preventing vascular leaky syndrome.

Background: Interleukin-2 (IL-2) is the first cancer therapeutic agent with an immunomodulatory function. Although it has been experimentally proven to be effective against metastatic renal cell carcinoma and metastatic melanoma, the clinical application of high-dose IL-2 (HDIL-2) has been limited because of its short half-life and severe side effects, such as vascular leakage syndrome (VLS) or capillary leaky syndrome (CLS). However, methods for overcoming this issue have not yet been identified. Methods: We discovered CU06-1004, an endothelial dysfunction blocker, through a previous study, and co-treated with IL-2 immunotherapy to confirm its inhibitory effect on HDIL-2-induced endothelial permeability. CU06-1004 was co-administered with HDIL-2 for 4 days in an in vivo mouse model. After drug injection, the mice were sacrificed, and Evans blue staining was performed. Results: In vitro, HDIL-2 treatment decreased HUVEC stability, which was rescued by co-treatment with CU06-1004. In our mouse model, co-administration of CU06-1004 and HDIL-2 prevented HDIL-2-induced vascular leakage by normalizing endothelial cells. Notably, the HDIL-2 and CU06-1004 combination therapy considerably reduced tumor growth in the B16F10 melanoma mouse model. Conclusion: Our data suggest that CU06-1004 acts as a potential anticancer drug candidate, not only by preventing HDIL-2-induced VLS but also by enhancing the anticancer effects of HDIL-2 immunotherapy.

Hybrids of carbonic anhydrase and cyclooxygenase inhibitors attenuate cardiac hypoxic inflammatory injuries.

Cardiac inflammation is easily accompanied by hypoxia, while hypoxia-induced injury and microenvironmental variations limit the efficacy of common anti-inflammatory drugs. In order to effectively attenuate myocardial injury caused by hypoxic and inflammatory injury, we designed and synthesized a kind of anti-inflammatory compounds by coupling cyclooxygenase (COX) and carbonic anhydrase (CA) inhibitors, and evaluated the activity and their mechanism in vitro and in vivo. It was found that these compounds were structurally stable and had two enzymatic inhibition activities. By inhibiting the activity of overexpressed CA under hypoxia, the acidic microenvironment can be regulated to inhibit the hypoxic injury, in which the pH-dependent primary drug resistance can be overcome to improve the anti-inflammatory effect of the COX inhibitor. Consequently, this study provides a new strategy for the treatment of cardiac inflammation accompanied by hypoxia.

The Efficiency and Safety of Mirabegron Monotherapy for the Treatment of Urge Incontinence in Women Aged >80 Years.

Objective We aimed to evaluate the efficacy and safety of mirabegron monotherapy in very older (>80 years) women with overactive bladder (OAB) who were discontinued anticholinergic drugs by the other departments. Material and methods The present retrospective study evaluated very older (>80 years) women with OAB who were discontinued anticholinergic drugs by the other departments between May 2018 and January 2021. Efficacy assessments were performed using Overactive Bladder-Validated Eight-Question (OAB-V8) scores before and after mirabegron monotherapy (12 weeks). Safety was evaluated based on adverse events (hypertension, nasopharyngitis, and urinary tract infection), electrocardiography, hypertension measure, uroflowmetry (UFM), and post-voiding. Patient data including demographic characteristics, diagnoses, values before and after mirabegron monotherapy, and adverse events were evaluated. Results A total of 42 very older (>80 years) women with OAB who used mirabegron monotherapy (50 mg per day) were included in this study. Frequency, nocturia, urgency, and total OAB-V8 scores were significantly lower after mirabegron monotherapy than before mirabegron monotherapy (p < 0.05, p < 0.05, p < 0.05, and p < 0.05, respectively). There was no significant difference between systolic-diastolic blood pressure and heart rate before and after mirabegron monotherapy treatment. Conclusion Mirabegron monotherapy is an effective and safe therapy in very older (>80 years) women with OAB.

DSN-DDI: an accurate and generalized framework for drug-drug interaction prediction by dual-view representation learning.

Drug-drug interaction (DDI) prediction identifies interactions of drug combinations in which the adverse side effects caused by the physicochemical incompatibility have attracted much attention. Previous studies usually model drug information from single or dual views of the whole drug molecules but ignore the detailed interactions among atoms, which leads to incomplete and noisy information and limits the accuracy of DDI prediction. In this work, we propose a novel dual-view drug representation learning network for DDI prediction ('DSN-DDI'), which employs local and global representation learning modules iteratively and learns drug substructures from the single drug ('intra-view') and the drug pair ('inter-view') simultaneously. Comprehensive evaluations demonstrate that DSN-DDI significantly improved performance on DDI prediction for the existing drugs by achieving a relatively improved accuracy of 13.01% and an over 99% accuracy under the transductive setting. More importantly, DSN-DDI achieves a relatively improved accuracy of 7.07% to unseen drugs and shows the usefulness for real-world DDI applications. Finally, DSN-DDI exhibits good transferability on synergistic drug combination prediction and thus can serve as a generalized framework in the drug discovery field.

Aptamer-Based Tumor-Targeted Diagnosis and Drug Delivery.

Early cancer identification is crucial for providing patients with safe and timely therapy. Highly dependable and adaptive technologies will be required to detect the presence of biological markers for cancer at very low levels in the early stages of tumor formation. These techniques have been shown to be beneficial in encouraging patients to develop early intervention plans, which could lead to an increase in the overall survival rate of cancer patients. Targeted drug delivery (TDD) using aptamer is promising due to its favorable properties. Aptamer is suitable for superior TDD system candidates due to its desirable properties including a high binding affinity and specificity, a low immunogenicity, and a chemical composition that can be simply changed.Due to these properties, aptamer-based TDD application has limited drug side effect along with organ damages. The development of aptasensor has been promising in TDD for cancer cell treatment. There are biomarkers and expressed molecules during cancer cell development; however, only few are addressed in aptamer detection study of those molecules. Its great potential of attachment of binding to specific target molecule made aptamer a reliable recognition element. Because of their unique physical, chemical, and biological features, aptamers have a lot of potential in cancer precision medicine.In this review, we summarized aptamer technology and its application in cancer. This includes advantages properties of aptamer technology over other molecules were thoroughly discussed. In addition, we have also elaborated the application of aptamer as a direct therapeutic function and as a targeted drug delivery molecule (aptasensor) in cancer cells with several examples in preclinical and clinical trials.

Bilateral, sequential orbital inflammatory syndrome associated with ruxolitinib.

Ruxolitinib is an oral Janus associated kinase (JAK) inhibitor commonly used in the treatment of primary myelofibrosis. We describe a case of an 86-year-old woman on ruxolitinib for primary myelofibrosis who presented with a three-day history of worsening left eyelid swelling, pain, and decreased vision. Her exam was notable for left upper lid ptosis, periorbital edema, and nearly complete external ophthalmoplegia along with diffuse conjunctival injection and microcystic corneal edema. An orbital computed tomography demonstrated left proptosis accompanied by extensive inflammatory changes of the preseptal and orbital soft tissues. She was diagnosed with acute left orbital inflammatory syndrome (OIS) and treated with intravenous methylprednisolone, one gram over 48 hours followed by an oral steroid taper as well as discontinuation of her ruxolitinib. Complete recovery was noted at her one-week follow-up visit and ruxolitinib was restarted. However, 3 weeks later, she presented with new right periorbital swelling and pain and was found to have a sequential right OIS for which she was again treated with methylprednisolone and discontinuation of ruxolitinib with the goal of transitioning to an alternative biologic agent. Although there are no previous documented cases of ruxolitinib associated orbital inflammatory syndrome, a similar JAK inhibitor medication, Fedratinib, has been reported to cause a similar side effect. We propose that her ruxolitinib paradoxically lead to a pro-inflammatory state leading to bilateral, sequential orbital inflammatory syndrome.

Identification of drug-side effect association via restricted Boltzmann machines with penalized term.

In the entire life cycle of drug development, the side effect is one of the major failure factors. Severe side effects of drugs that go undetected until the post-marketing stage leads to around two million patient morbidities every year in the United States. Therefore, there is an urgent need for a method to predict side effects of approved drugs and new drugs. Following this need, we present a new predictor for finding side effects of drugs. Firstly, multiple similarity matrices are constructed based on the association profile feature and drug chemical structure information. Secondly, these similarity matrices are integrated by Centered Kernel Alignment-based Multiple Kernel Learning algorithm. Then, Weighted K nearest known neighbors is utilized to complement the adjacency matrix. Next, we construct Restricted Boltzmann machines (RBM) in drug space and side effect space, respectively, and apply a penalized maximum likelihood approach to train model. At last, the average decision rule was adopted to integrate predictions from RBMs. Comparison results and case studies demonstrate, with four benchmark datasets, that our method can give a more accurate and reliable prediction result.

Rhabdomyolysis Caused by Isotretinoin and Exercise in an Otherwise Healthy Female Patient.

Acne vulgaris is one of the most common skin conditions treated by healthcare providers. Isotretinoin is a well-known and effective treatment for nodulocystic and scarring acne. Rarely, and usually in combination with exercise, patients treated with isotretinoin can develop rhabdomyolysis, a potentially life-threatening breakdown of muscle associated with elevated creatine kinase (CK). Here, we report a rare case of a female patient developing rhabdomyolysis three months after starting exercise and isotretinoin. She was treated with supportive care and medication was discontinued, resulting in a full recovery. Careful inquiry into the patient's exercise habits, along with a thorough review of systems at each visit can help identify high-risk patients. Routine monitoring of liver enzymes, specifically aspartate aminotransferase elevations, may provide a prompt to check a patient's CK. Though regular monitoring of CK is not currently recommended, given the prevalence of regular exercise in certain patient populations, this case reinforces the importance of counseling patients on this potential side effect.

idse-HE: Hybrid embedding graph neural network for drug side effects prediction.

In drug development, unexpected side effects are the main reason for the failure of candidate drug trials. Discovering potential side effects of drugsin silicocan improve the success rate of drug screening. However, most previous works extracted and utilized an effective representation of drugs from a single perspective. These methods merely considered the topological information of drug in the biological entity network, or combined the association information (e.g. knowledge graph KG) between drug and other biomarkers, or only used the chemical structure or sequence information of drug. Consequently, to jointly learn drug features from both the macroscopic biological network and the microscopic drug molecules. We propose a hybrid embedding graph neural network model named idse-HE, which integrates graph embedding module and node embedding module. idse-HE can fuse the drug chemical structure information, the drug substructure sequence information and the drug network topology information. Our model deems the final representation of drugs and side effects as two implicit factors to reconstruct the original matrix and predicts the potential side effects of drugs. In the robustness experiment, idse-HE shows stable performance in all indicators. We reproduce the baselines under the same conditions, and the experimental results indicate that idse-HE is superior to other advanced methods. Finally, we also collect evidence to confirm several real drug side effect pairs in the predicted results, which were previously regarded as negative samples. More detailed information, scientific researchers can access the user-friendly web-server of idse-HE at http://bioinfo.jcu.edu.cn/idse-HE. In this server, users can obtain the original data and source code, and will be guided to reproduce the model results.

Integrating specific and common topologies of heterogeneous graphs and pairwise attributes for drug-related side effect prediction.

MOTIVATION: Computerized methods for drug-related side effect identification can help reduce costs and speed up drug development. Multisource data about drug and side effects are widely used to predict potential drug-related side effects. Heterogeneous graphs are commonly used to associate multisourced data of drugs and side effects which can reflect similarities of the drugs from different perspectives. Effective integration and formulation of diverse similarities, however, are challenging. In addition, the specific topology of each heterogeneous graph and the common topology of multiple graphs are neglected. RESULTS: We propose a drug-side effect association prediction model, GCRS, to encode and integrate specific topologies, common topologies and pairwise attributes of drugs and side effects. First, multiple drug-side effect heterogeneous graphs are constructed using various kinds of similarities and associations related to drugs and side effects. As each heterogeneous graph has its specific topology, we establish separate module based on graph convolutional autoencoder (GCA) to learn the particular topology representation of each drug node and each side effect node, respectively. Since multiple graphs reflect the complex relationships among the drug and side effect nodes and contain common topologies, we construct a module based on GCA with sharing parameters to learn the common topology representations of each node. Afterwards, we design an attention mechanism to obtain more informative topology representations at the representation level. Finally, multi-layer convolutional neural networks with attribute-level attention are constructed to deeply integrate the similarity and association attributes of a pair of drug-side effect nodes. Comprehensive experiments show that GCRS's prediction performance is superior to other comparing state-of-the-art methods for predicting drug-side effect associations. The recall rates in top-ranked candidates and case studies on five drugs further demonstrate GCRS's ability in discovering potential drug-related side effects. CONTACT: zhang@hlju.edu.cn.