A novel method of manual positioning based on anatomical mark (shoulder-to-shoulder) to prevent postoperative leg-length discrepancy for femoral neck fractures in hip arthroplasty

OBJECTIVE

To determine whether the two lower extremities are of equal length after hip arthroplasty for femoral neck fractures, we developed a novel method of manual positioning based on anatomical mark (shoulder-to-shoulder) in hip arthroplasty.

METHODS

Patients with femoral neck fractures requiring hip arthroplasty from July 2020 to March 2022 in the orthopedic department of Jinjiang Municipal Hospital, Fujian Province, China were recruited. Hip arthroplasty was performed using the proposed "shoulder-to-shoulder" method of manual positioning based on anatomical mark in 52 patients with femoral neck fractures who met the inclusion criteria. "Shoulder-to-shoulder" was achieved by alignment of the marked femoral "shoulder" and the "shoulder" of prosthesis stem. There were 16 male and 36 female patients, with 27 undergoing total hip arthroplasty (THA) and 25 undergoing hip hemiarthroplasty (HA). The fractures were categorized according to the Garden classification: type II, type III, and type IV in 5, 11, and 36 patients, respectively. The vertical distance from the apex of the medial margin of the femoral trochanter to the tear drop line on both sides which was regarded as the length of both limbs were compared via postoperative imaging, and the apex-shoulder distance on the ipsilateral side measured via postoperative imaging was compared with those measured intraoperatively.

RESULTS

All patients completed the surgery successfully. The measurement results for the lower extremities after THA were as follows: contralateral group, 43.87 ± 5.59 mm; ipsilateral group, 44.64 ± 5.43 mm. The measurement results for the lower extremities after HA were as follows: contralateral group, 45.18 ± 7.82 mm; ipsilateral group, 45.16 ± 6.43 mm. The measurement results for the lower extremities after all arthroplasties were as follows: contralateral group, 44.50 ± 6.72 mm; ipsilateral group, 44.89 ± 5.90 mm. The results for the apex-shoulder distance were as follows: postoperative imaging, 19.44 ± 3.54 mm; intraoperative apex-shoulder distance, 27.28 ± 2.84 mm. Statistical analysis results indicated no statistically significant difference in the postoperative bilateral lower extremity length after hip arthroplasty (P = 0.75), while a statistically significant difference was found between the intraoperative and postoperative imaging measurements of the apex-shoulder distance (P < 0.01).

CONCLUSION

The novel method of manual positioning based on anatomical mark (shoulder-to-shoulder) for femoral neck fractures in hip arthroplasty is simple and accurate, making it effective for preventing postoperative bilateral leg length discrepancy.

[The application value of cardiac magnetic resonance quantitative T1 mapping technique for risk stratification in patients with pulmonary arterial hypertension]

Objective: To explore the application value of cardiac magnetic resonance (CMR) native T₁ mapping for risk stratification in patients with pulmonary arterial hypertension (PAH). Methods: A total of 59 patients with diagnosed PAH and clear-documented risk status in Peking Union Medical College Hospital and underwent CMR examination between January 2019 and December 2021 were retrospectively included, which including 12 males and 47 females, aged from 4 to 77 (31±13) years. Those patients were subdivided into two groups based on the clinically-assessed risk status: low-risk group (n=30) and intermediate-/high-risk group (n=29). Twenty-five healthy individuals were included as controls. Base, midventricular, and apical inferior right ventricular insertion point (IRVIP) native T₁ values on short axis images were measured. Native T1 values in PAH patients and control group, in low-risk group and intermediate-/high-risk group were compared, respectively, and receiver operating characteristics (ROC) curves with area under the curves (AUC) were calculated to evaluate the application value of native T₁ values for risk stratification in PAH patients. Results: Base, midventricular and apical IRVIP native T₁ of PAH patients were all significantly increased as compared to controls [Base:(1 439.31±129.96) vs (1 282.36±37.18) ms;midventricular:(1 450.32±111.55) vs (1 287.56±53.16) ms;apical:(1 444.12±109.15) vs (1 266.36±75.31) ms](all P<0.001). The midventricular IRVIP native T₁ values were significantly higher in patients in intermediate-/high-risk status as compared to those in low-risk status [ (1 493.24±126.32) vs (1 428.50±85.73) ms,P=0.026]. The AUC of mid ventricle IRVIP native T₁ for distinguishing patients in intermediate-/high-risk status was 0.741. The base [(1 458.21±134.96) vs (1 421.03±104.75) ms, P=0.241] and apical [(1 465.90±125.36) vs (1 423.07±87.87) ms,P=0.136] IRVIP native T1 values in patients in intermediate-/high-risk group were also numerically higher as compared with patients in low-risk status, however, without statistical significant (both P>0.05). Conclusion: Midventricular IRVIP native T₁ value might have a role for assisting in risk stratification in PAH patients, which was clinically significant for facilitating the work-up and prognosis improvement of PAH patients.

Adversarial Reinforced Instruction Attacker for Robust Vision-Language Navigation

Language instruction plays an essential role in the natural language grounded navigation tasks. However, navigators trained with limited human-annotated instructions may have difficulties in accurately capturing key information from the complicated instruction at different timesteps, leading to poor navigation performance. In this paper, we exploit to train a more robust navigator which is capable of dynamically extracting crucial factors from the long instruction, by using an adversarial attacking paradigm. Specifically, we propose a Dynamic Reinforced Instruction Attacker (DR-Attacker), which learns to mislead the navigator to move to the wrong target by destroying the most instructive information in instructions at different timesteps. By formulating the perturbation generation as a Markov Decision Process, DR-Attacker is optimized by the reinforcement learning algorithm to generate perturbed instructions sequentially during the navigation, according to a learnable attack score. Then, the perturbed instructions, which serve as hard samples, are used for improving the robustness of the navigator with an effective adversarial training strategy and an auxiliary self-supervised reasoning task. Experimental results on both Vision-and-Language Navigation (VLN) and Navigation from Dialog History (NDH) tasks show the superiority of our proposed method over state-of-the-art methods. Moreover, the visualization analysis shows the effectiveness of the proposed DR-Attacker, which can successfully attack crucial information in the instructions at different timesteps. Code is available at https://github.com/expectorlin/DR-Attacker.

Challenge and Strategies in Room Temperature Sodium-Sulfur Batteries: A Comparison with Lithium-Sulfur Batteries

Metal-sulfur batteries exhibit great potential as next-generation rechargeable batteries due to the low sulfur cost and high theoretical energy density. Sodium-sulfur (Na-S) batteries present higher feasibility of long-term development than lithium-sulfur (Li-S) batteries in technoeconomic and geopolitical terms. Both lithium and sodium are alkali metal elements with body-centered cubic structures, leading to similar physical and chemical properties and exposing similar issues when employed as the anode in metal-sulfur batteries. Indeed, some inspiration for mechanism researches and strategies in Na-S systems comes from the more mature Li-S systems. However, the dissimilarities in microscopic characteristics determine that Na-S is not a direct Li-S analogue. Herein, the daunting challenges derived by the differences of fundamental characteristics in Na-S and Li-S systems are discussed. And the corresponding strategies in Na-S batteries are reviewed. Finally, general conclusions and perspectives toward the research direction are presented based on the dissimilarities between both systems. This review attempts to provide important insights to facilitate the assimilation of the available knowledge on Li-S systems for accelerating the development of Na-S batteries on the basis of their dissimilarities.

Joint Learning of Neural Transfer and Architecture Adaptation for Image Recognition

Current state-of-the-art visual recognition systems usually rely on the following pipeline: 1) pretraining a neural network on a large-scale data set (e.g., ImageNet) and 2) finetuning the network weights on a smaller, task-specific data set. Such a pipeline assumes that the sole weight adaptation is able to transfer the network capability from one domain to another domain based on a strong assumption that a fixed architecture is appropriate for all domains. However, each domain with a distinct recognition target may need different levels/paths of feature hierarchy, where some neurons may become redundant, and some others are reactivated to form new network structures. In this work, we prove that dynamically adapting network architectures tailored for each domain task along with weight finetuning benefits in both efficiency and effectiveness, compared to the existing image recognition pipeline that only tunes the weights regardless of the architecture. Our method can be easily generalized to an unsupervised paradigm by replacing supernet training with self-supervised learning in the source domain tasks and performing linear evaluation in the downstream tasks. This further improves the search efficiency of our method. Moreover, we also provide principled and empirical analysis to explain why our approach works by investigating the ineffectiveness of existing neural architecture search. We find that preserving the joint distribution of the network architecture and weights is of importance. This analysis not only benefits image recognition but also provides insights for crafting neural networks. Experiments on five representative image recognition tasks, such as person re-identification, age estimation, gender recognition, image classification, and unsupervised domain adaptation, demonstrate the effectiveness of our method.

Impacts of menopause hormone therapy on mood disorders among postmenopausal women

OBJECTIVE

This study aimed to explore the modulatory effects of menopause hormone therapy (MHT) on mood disorders among postmenopausal women.

METHODS

A cross-sectional study was conducted to recruit postmenopausal women, including patients (arranged MHT for over 3 years as the medication group) and non-MHT controls. All participants were asked to respond to the Center for Epidemiological Studies Depression Scale (CES-D) and Generalized Anxiety Disorder Screener (GAD-7) questionnaires to assess their depression and anxiety status.

RESULTS

A total of 230 cases from the two groups were determined based on propensity score matching analysis by matching the menopausal age and menopausal durations. We found that MHT served as a favorable modulator in the depression status of postmenopausal women. Among the four factors of the CES-D questionnaire, our data indicated that the differences between the two groups fell primarily into two aspects: depressive emotion, and somatic symptoms or retarded activities. MHT was mainly involved in improving the depression of overweight women. However, no substantial effects of MHT were observed on the regulation of anxiety.

CONCLUSION

Postmenopausal women, especially the overweight population, who have experienced MHT exhibited an improved depressive status but not their anxiety condition.

H2S regulation of ferroptosis attenuates sepsis‑induced cardiomyopathy

Ferroptosis is a non-apoptotic form of cell death mediated by reactive oxygen species (ROS). Iron metabolism disorders play a key role in sepsis-induced cardiomyopathy (SIC). While hydrogen sulfide (H₂S) inhibits SIC, it is unknown if it does so by controlling ferroptosis. The present study evaluated whether sodium hydrosulfide (NaHS), an H₂S donor, alleviates SIC by decreasing ferroptosis. Lipopolysaccharide (LPS) was employed to induce an in vitro model of septic myocardial injury in rat H9c2 cardiomyocytes. The myocardial injury model of septic rats was established by cecal ligation and puncture (CLP). Cardiomyocyte injury was evaluated using Cell Counting Kit-8 and myocardial enzyme assay and hematoxylin and eosin (H&E) staining. The cardiac function of rats was assessed using echocardiography and changes in myocardial fibers and mitochondria were evaluated using H&E staining and transmission electron microscopy, respectively. Fe²⁺, glutathione and malondialdehyde levels in cardiomyocytes were detected using assay kits, ROS and mitochondrial membrane potential changes were detected using fluorescent probes and ferroptosis and Beclin 1 (BECN1) signaling pathway-associated protein expression levels were semi-quantified using western blotting. NaHS decreased ferroptosis of H9c2 cells induced by LPS and decreased injury of myocardial cells by improving iron metabolism disorder and oxidative stress levels. Furthermore, in vivo results demonstrated that NaHS attenuated CLP-induced septic myocardial ferroptosis and significantly improved cardiac dysfunction in septic rats compared with the CLP group. NaHS was demonstrated to attenuate sepsis-induced myocardial cell and tissue injury by significantly inhibiting the phosphorylation of BECN1 and significantly increasing expression levels of the ferroptosis regulatory proteins solute carrier family 7 member 11 and glutathione peroxidase 4. The results of the present study suggested that by regulating the BECN1 signaling pathway, NaHS may decrease the incidence of myocardial ferroptosis, thereby improving SIC.

Abstract PO001: Preclinical activity of glypican-3 (GPC3) and NKp46 directed FLEX-NK™ engager antibody (CYT-303) in combination with iPSC derived natural killer cells (iNKs) or peripheral blood (PB) NK cells in hepatocellular carcinoma (HCC)

BACKGROUND: GPC3 is highly expressed in HCC while it is hardly expressed in adult normal tissues except placenta. Given the presence of dysfunctional NK cells in the HCC tumor microenvironment that contribute to poor anti-tumor immune responses, we sought to develop an iNK cell therapy in combination with GPC3 targeted NK cell engagers (NKE). FLEX-NK™ is a proprietary tetravalent IgG1-like multifunctional antibody NKE platform with a novel FLEX-linker to allow for simultaneous binding to different antigens on different cells. CYT-303 was constructed to engage NK cells via the activation receptor NKp46 and to target cancer cells via GPC3. We hypothesized that CYT-303 may be active in the tumor microenvironment where there are sufficient number and function of endogenous NK cells, but that this activity may be augmented by fresh functional NK cells. iNK cells derived from iPSC, a uniform starting material with unlimited self-renewal capabilities, can be expanded to produce a universal off-the-shelf allogeneic therapy. We investigated the combination of CYT-303 to enhance tumor specificity and potency of iNKs. METHODS: CYT-303 was expressed in CHOZEN cells and purified by step column chromatography. iNK’s were differentiated from iPSC’s and expanded using irradiated feeder cells. PB-NK cells were purified from healthy donor blood. iNK and PB-NK cytotoxicity against Hep3B tumors was evaluated in the presence of CYT-303 or isotype control antibodies at a fixed E/T ratio. CYT-303 pharmacokinetics in mice was evaluated following a single intravenous injection. Anti-tumor efficacy of the iNK combination with CYT-303 was evaluated in NSG-IL15 mice bearing subcutaneous Hep3B tumors following intra-tumoral injection of iNKs and intravenous dosing with CYT-303. Blood alpha-fetoprotein (AFP) levels were assessed by immunoassay. Cytokine release assessments for CYT-303 was evaluated in an in-vitro PBMC activation assay. RESULTS: iNK cells express high levels of multiple activation receptors including NKp46, but low levels of CD16. Despite low expression of CD16, natural cytotoxicity of iNK’s against Hep3B tumors was higher than with fresh PB-NK and CYT-303 redirected killing of Hep3B tumors was comparable to PB-NK’s. The CYT-303 pharmacokinetics study in mice showed a terminal half-life of 17 hrs. In a Hep3B tumor model in NSG-hIL-15 mice, the combination of CYT-303 and iNK’s showed significantly greater tumor growth inhibition compared to iNK’s alone plus an IgG1 isotype control. Blood AFP levels decreased in the CYT-303 plus iNK combination compared to iNK’s alone. Cytokine release assessment of CYT-303 in the human PBMC assay showed no evidence of cytokine release while high levels of cytokine release was observed with anti-CD28 (TGN1412) and CD3 antibody controls. CONCLUSIONS: The combination of NK cell engager CYT-303 and iNK cells is active in preclinical models of HCC with low levels of cytokine release. These data support further development of CYT-303 and iNKs to engage and empower NK cells as an off the shelf therapeutic approach in HCC.

Citation Format: Antonio Arulanandam, Liang Lin, Hao-Ming Chang, Harish Potu, Vishal Khairnar, David Zou, Melissa Triggiano, Nejmi Dilmac, Yinan Yang, Shira Kahlon, Stanley Frankel, Jean Kadouche, Daniel Teper, Ofer Mandelboim, Yoav Hershkovitz, Yaron Ilan, Wei Li. Preclinical activity of glypican-3 (GPC3) and NKp46 directed FLEX-NK™ engager antibody (CYT-303) in combination with iPSC derived natural killer cells (iNKs) or peripheral blood (PB) NK cells in hepatocellular carcinoma (HCC) [abstract]. In: Proceedings of the AACR Special Conference: Advances in the Pathogenesis and Molecular Therapies of Liver Cancer; 2022 May 5-8; Boston, MA. Philadelphia (PA): AACR; Clin Cancer Res 2022;28(17_Suppl):Abstract nr PO001.

Adjustable Mixed Conductive Interphase for Dendrite-Free Lithium Metal Batteries

Lithium (Li) metal batteries with high energy density are of great promise for next-generation energy storage; however, they suffer from severe Li dendritic growth and an unstable solid electrolyte interphase. In this study, a mixed ionic and electronic conductive (MIEC) interphase layer with an adjustable ratio assembled by ZnO and Zn nanoparticles is developed. During the initial cycle, the in situ formed Li₂O with high ionic conductivity and a lithiophilic LiZn alloy with high electronic conductivity enable fast Li⁺ transportation in the interlayer and charge transfer at the ion/electron conductive junction, respectively. The optimized interface kinetics is achieved by balancing the ion migration and charge transfer in the MIEC Li₂O-LiZn interphase. As a result, the symmetric cell with MIEC interphase delivers superior cycling stability of over 1200 h. Also, Li||Zn-ZnO@PP||LFP (LFP = LiFePO₄) full cells exhibit long cyclic life for 2000 cycles with a very high capacity retention of 91.5% at a high rate of 5 C and stable cycling for 350 cycles at a high LFP loading mass of 13.27 mg cm^-2.

Predicting recurrence and metastasis risk of endometrial carcinoma via prognostic signatures identified from multi-omics data

Objectives

Endometrial carcinoma (EC) is one of the three major gynecological malignancies, in which 15% - 20% patients will have recurrence and metastasis. Though there are many studies on the prognosis on this cancer, the performances of existing models evaluating the risk of its recurrence and metastasis are yet to be improved. In addition, a comprehensive multi-omics analyses on the prognostic signatures of EC are on demand. In this study, we aimed to construct a relatively stable and reliable model for predicting recurrence and metastasis of EC. This will help determine the risk level of patients and choose appropriate adjuvant therapy, thereby avoiding improper treatment, and improving the prognosis of patients.

Methods

The mRNA, microRNA (miRNA), long non-coding RNA (lncRNA), copy number variation (CNV) data and clinical information of patients with EC were downloaded from The Cancer Genome Atlas (TCGA). Differential expression analyses were performed between the recurrence or metastasis group and the non-recurrence/metastasis group. Then, we screened potential prognostic markers from the four kinds of omics data respectively and established prediction models using three classifiers.

Results

We achieved differential expressed mRNAs, lncRNAs, miRNAs and CNVs between the two groups. According to feature selection scores by the random forest algorithm, 275 CNV features, 50 lncRNA features, 150 miRNA features and 150 mRNA features were selected, respectively. And the prediction model constructed by the features of lncRNA data using random forest method showed the best performance, with an area under the curve of 0.763, and an accuracy of 0.819 under 10-fold cross-validation.

Conclusion

We developed a computational model using omics information, which is able to predicting recurrence and metastasis risk of EC accurately.

Molecular classification grade 3 endometrial endometrioid carcinoma using a next-generation sequencing–based gene panel

Over the past two decades, the incidence of endometrial cancer (EC) is increasing, and there is a need for molecular biomarkers to predict prognosis and guide treatment. A recent study from The Cancer Genome Atlas suggested to implement the EC analysis by molecular profile for improving diagnosis, prognosis, and therapeutic treatment. In this study, next-generation sequencing was performed on 70 cases of G3 endometrioid ECs (EECs) using an 11-gene panel (TP53, MLH1, MSH2, MSH6, PMS2, EPCAM, PIK3CA, CTNNB1, KRAS, PTEN, and POL) for molecular classification. The molecular classification based on the 11-gene NGS panel identified four molecular subgroups: POLE-ultramutated (n = 20, 28.6%), MSI-H (n = 27, 38.6%), NSMP (n = 13, 18.6%) and TP53mut (n = 10, 14.3%). The NGS method showed 98.6% (69 of 70 cases, kappa value 98%) in concordance with the cases assessed by immunohistochemistry (IHC). Among the seven dead cases, four were MSI-H tumors, two were TP53mut/p53abn tumors, and one was NSMP tumors with an average overall survival (OS) of 14.7 months. TP53mut subgroup showed that poor OS rates and POLE group have favorable prognosis. Our work suggested that the 11-gene panel is suitable for molecular classification in G3 EECs and for guiding prognosis and treatment decisions.

Integrative lipidomic features identify plasma lipid signatures in chronic urticaria

Chronic urticaria (CU) is a chronic inflammatory skin disease mainly mediated by mast cells. Lipids exert essential functions in biological processes; however, the role of lipids in CU remains unclear. Nontargeted lipidomics was performed to investigate the differential lipid profiles between CU patients and healthy control (HC) subjects. Functional validation studies were performed in vitro and in vivo including β-hexosaminidase release examination from mast cells and passive cutaneous anaphylaxis (PCA) mouse model. We detected dramatically altered glycerophospholipids in CU patients compared with HCs. Phosphatidylserine (PS), phosphatidylethanolamine (PE), and phosphatidylglycerol (PG) were increased, while phosphatidylcholine (PC) was reduced in CU patients. The reduction in PC was related to a high weekly urticaria activity score (UAS7), while PS was positively associated with the dermatology life quality index (DLQI). We also identified the differential lipid profiles between chronic spontaneous urticaria (CSU), symptomatic dermographism (SD), and CSU coexist with SD. CU patients were classified into two subtypes (subtype 1 and subtype 2) based on consensus clustering of lipid profiling. Compared with patients in subtype 2, patients in subtype 1 had elevated levels of PC (18:0e/18:2) and PE (38:2), and lower urticaria control test (UCT) scores indicated worse clinical efficiency of secondary generation H1 antihistamines treatment. Importantly, we found that supplementation with PC could attenuate IgE-induced immune responses in mast cells. In general, We described the landscape of plasma lipid alterations in CU patients and provided novel insights into the role of PC in mast cells.

Wild Panax plants adapt to their thermal environment by harboring abundant beneficial seed endophytic bacteria

The seed microbiome of crop wild relatives is a potential reservoir of beneficial traits that potentially improve their host plant resilience to fluctuating environments and pathogenic threats. Herein, we studied the seed microbiome of three species of the medicinal genus Panax (P. vietnamensis, P. japonicas, and P. stipuleanatus) collected from seven locations in Southwest China. We used qPCR and metabarcoding high-throughput sequencing to target both endophytic bacteria and fungi. Seed bacterial absolute abundance (1.1 × 109∼1.0 × 107 gene copy numbers per gram seed) was substantially higher than that of fungi (7.6 × 105∼3.7 × 102). Host plant genotype was the main driver of seed microbiome composition for both bacteria and fungi. Panax growing hypothermal environments significantly shaped their seed endophytic bacterial but not fungal microbiota. The three Panax species’ seeds harbored unique microbes [averaged ∼150 amplicon sequence variants (ASVs)], sharing only 12 bacterial ASVs (half affiliated to Halomonas) and four fungal ASVs. Network analysis showed that the Panax seed endophytic bacteria tend to form inter-weaved functional modules that are majorly connected by core members from the genus Halomonas, Pseudomonas, and Pantoea. These genera have been associated with nutrient cycling, plant, disease suppression, and tolerance to environmental fluctuation. Together, these novel insights may shade light on the ecological strategies of wild Panax plants adaptation to their thermal environment by possessing abundant beneficial seed endophytic bacteria.

[Analysis of the efficacy and related influencing factors of pelvic packing in the treatment of intractable postpartum hemorrhage after emergency perinatal hysterectomy]

Objective: To investigate the effect of pelvic packing on the control of intractable postpartum hemorrhage after emergency perinatal hysterectomy (EPH). Methods: Eleven cases with complete clinical data of pelvic packing due to failure of hemostasis after EPH were collected to evaluate the outcome, complications, hospital stay of pregnant women, and to analyze the factors affecting the effect of pelvic packing. The cases included patients who were admitted to the Third Affiliated Hospital of Guangzhou Medical University after pelvic packing treatment in the other hospital due to continuous bleeding after EPH or who were referred to our hospital for pelvic packing treatment due to continuous bleeding after EPH from January 2014 to August 2021. Results: The median gestational week of 11 pregnant women was 38.3 weeks(38.0-39.9 weeks) , and the methods of termination of pregnancy were cesarean section in 7 cases (7/11) and vaginal delivery in 4 cases (4/11). The median time between postpartum hemorrhage and pelvic tamponade was 10 hours (5-57 hours), the median amount of bleeding was 8 500 ml(4 800-15 600 ml) , the median number of pelvic tamponade was 3 pieces (2-7 pieces), and the median retention time of gauze pad was 6.0 days (3.0-6.0 days). The median frequency of laparotomy in this pregnancy was 3 times (2-3 times), with a maximum of 4 among the 11 cases, the first pelvic packing was successful in hemostasis in 9 cases, and the final successful treatment in all of the 11 cases. All parturients had hemorrhagic shock (11/11) and disseminated intravascular coagulation (11/11) before pelvic packing. Other common complications were multiple organ dysfunction syndrome (9/11), cardiac arrest (4/11), deep vein thrombosis (3/11), septic shock (3/11), and intestinal obstruction (1/11). All parturients took out the gauze after the coagulation function returned to normal and there was no active bleeding. The recovery time of coagulation function in 11 cases was 3 days (3-5 days), the retention time of gauze pad was 6 days (3-6 days), the median length of stay in intensive care unit was 14 days (11-26 days), and the median total length of stay was 22 days (16-49 days). Conclusions: Pelvic packing could be used as a temporary strategy for intractable postpartum hemorrhage after EPH, which provides a key time for injury control resuscitation for patients with unstable vital signs. This technology provides an opportunity for referral to superior medical institutions and further treatment.

Knowledge-Routed Visual Question Reasoning: Challenges for Deep Representation Embedding

Though beneficial for encouraging the visual question answering (VQA) models to discover the underlying knowledge by exploiting the input-output correlation beyond image and text contexts, the existing knowledge VQA data sets are mostly annotated in a crowdsource way, e.g., collecting questions and external reasons from different users via the Internet. In addition to the challenge of knowledge reasoning, how to deal with the annotator bias also remains unsolved, which often leads to superficial overfitted correlations between questions and answers. To address this issue, we propose a novel data set named knowledge-routed visual question reasoning for VQA model evaluation. Considering that a desirable VQA model should correctly perceive the image context, understand the question, and incorporate its learned knowledge, our proposed data set aims to cut off the shortcut learning exploited by the current deep embedding models and push the research boundary of the knowledge-based visual question reasoning. Specifically, we generate the question-answer pair based on both the visual genome scene graph and an external knowledge base with controlled programs to disentangle the knowledge from other biases. The programs can select one or two triplets from the scene graph or knowledge base to push multistep reasoning, avoid answer ambiguity, and balance the answer distribution. In contrast to the existing VQA data sets, we further imply the following two major constraints on the programs to incorporate knowledge reasoning. First, multiple knowledge triplets can be related to the question, but only one knowledge relates to the image object. This can enforce the VQA model to correctly perceive the image instead of guessing the knowledge based on the given question solely. Second, all questions are based on different knowledge, but the candidate answers are the same for both the training and test sets. We make the testing knowledge unused during training to evaluate whether a model can understand question words and handle unseen combinations. Extensive experiments with various baselines and state-of-the-art VQA models are conducted to demonstrate that there still exists a big gap between the model with and without groundtruth supporting triplets when given the embedded knowledge base. This reveals the weakness of the current deep embedding models on the knowledge reasoning problem.

Identification of NAD+ Metabolism-Derived Gene Signatures in Ovarian Cancer Prognosis and Immunotherapy

Background: Nicotinamide adenine dinucleotide (NAD⁺) has emerged as a critical regulator of cell signaling and survival pathways, affecting tumor initiation and progression. In this study it was investigated whether circulating NAD⁺ metabolism-related genes (NMRGs) could be used to predict immunotherapy response in ovarian cancer (OC) patients.

Method: In this study, NMRGs were comprehensively examined in OC patients, three distinct NMRGs subtypes were identified through unsupervised clustering, and an NAD⁺-related prognostic model was generated based on LASSO Cox regression analysis and generated a risk score (RS). ROC curves and an independent validation cohort were used to assess the model’s accuracy. A GSEA enrichment analysis was performed to investigate possible functional pathways. Furthermore, the role of RS in the tumor microenvironment, immunotherapy, and chemotherapy was also investigated.

Result: We found three different subgroups based on NMRGs expression patterns. Twelve genes were selected by LASSO regression to create a prognostic risk signature. High-RS was founded to be linked to a worse prognosis. In Ovarian Cancer Patients, RS is an independent prognostic marker. Immune infiltrating cells were considerably overexpressed in the low-RS group, as immune-related functional pathways were significantly enriched. Furthermore, immunotherapy prediction reveal that patients with low-RS are more sensitive to immunotherapy.

Conclusion: For a patient with OC, NMRGs are promising biomarkers. Our prognostic signature has potential predictive value for OC prognosis and immunotherapy response. The results of this study may help improve our understanding of NMRG in OCs.

Abstract 2752: Preclinical characterization of FLEX-NK࣪ tetravalent NKp46 engager directed against GPC3 (CYT-303) alone or in combination with iPSC derived Natural Killer cells (iNKs) against hepatocellular carcinoma (HCC)

Glypican-3 (GPC3) is highly expressed in multiple solid tumors including HCC while it is hardly expressed in adult normal tissues except placenta. GPC3 promotes Wnt-dependent cell proliferation, and its expression is correlated with poor prognosis in HCC. NK cells exhibit innate anti-tumor activity owing to the expression of multiple activating receptors, such as NKp46. NKp46 is expressed in all NK cells including tumor-infiltrating NK cells. FLEX-NKTrademark is a proprietary platform for production of tetravalent IgG1-like multifunctional antibody NK engagers with a novel FLEX-linker to allow for simultaneous binding of both the targeted cancer cells and NK cells. A novel humanized NKp46 binder that does not induce NKp46 internalization and a humanized GPC3 binder that targets the membrane-proximal lobe of the GPC3 were combined on the novel FLEX-NKTrademark scaffold to create the NK engager CYT-303. CYT-303 has higher affinity for GPC3 compared to NKp46 to skew binding preferentially to GPC3 expressing tumor cells prior to binding NK cells expressing NKp46 to optimize targeted killing. CYT-303 showed significantly higher dose dependent peripheral blood NK cell redirected cytotoxicity and degranulation against GPC3 expressing Hep3B tumor cells compared to GPC3 or NKp46 monoclonal antibodies alone suggesting that co-engagement of NKp46 and GPC3 via an immunological synapse is required for optimal tumor killing by CYT-303. Low NK cell numbers or suppression of NK cell function in the tumor microenvironment may limit the clinical activity of FLEX-NKTrademark engagers. iNK cells derived from iPSC, a uniform starting material with unlimited self-renewal capabilities, can be expanded to produce a universal off-the-shelf allogeneic therapy that can be used in combination with FLEX-NKTrademark engagers. We studied the efficacy of the combination of a FLEX-NKTrademark antibody and iNKs. The iNK cells express high levels of multiple activation receptors including NKp46 and showed good cytotoxic activity against HCC cell line Hep3B. The iNKs also showed anti-tumor activity in NSG-hIL15 mice bearing HCC subcutaneous tumors as demonstrated by the presence of CD56+CD3-, NKp46+, NKG2D+ iNKs in the tumors at day 21 post-intratumoral injection of the iNKs. CYT-303 greatly enhanced the cytotoxic activity of iNKs and cytolysis of Hep3B tumor cells in-vitro. CYT-303 and iNK cells, alone or in combination, demonstrate anti-tumor activity against HCC that warrants clinical development.

Citation Format: Antonio Arulanandam, Liang Lin, Hao-Ming Chang, David Zou, Melissa Triggiano, Nejmi Dilmac, Yinan Wang, Shira Kahlon, Stanley Frankel, Jean Kadouche, Daniel Teper, Ofer Mandelboim, Wei Li. Preclinical characterization of FLEX-NK࣪ tetravalent NKp46 engager directed against GPC3 (CYT-303) alone or in combination with iPSC derived Natural Killer cells (iNKs) against hepatocellular carcinoma (HCC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2752.

15-Month Health Outcomes and the Related Risk Factors of Hospitalized COVID-19 Patients From Onset: A Cohort Study

Objective

The long-term impact of COVID-19 on patient health has been a recent focus. This study aims to determine the persistent symptoms and psychological conditions of patients hospitalized with COVID-19 15 months after onset, that patients first developed symptoms. The potential risk factors were also explored.

Methods

A cohort of COVID-19 patients discharged from February 20, 2020 to March 31, 2020 was recruited. Follow-ups were conducted using validated questionnaires and psychological screening scales at 15 months after onset to evaluate the patients' health status. The risk factors for long-term health impacts and their associations with disease severity was analyzed.

Findings

534 COVID-19 patients were enrolled. The median age of the patients was 62.0 years old (IQR 52.0–70.0) and 295 were female (55.2%). The median time from onset to follow-up was 460.0 (451.0–467.0) days. Sleep disturbance (18.5%, 99/534) and fatigue (17.2%, 92/534) were the most common persistent symptoms. 6.4% (34/534) of the patients had depression, 9.2% (49/534) were anxious, 13.0% (70/534) had insomnia and 4.7% (25/534) suffered from post-traumatic stress disorder (PTSD). Multivariate adjusted logistic regression analysis showed that glucocorticoid use during hospitalization (OR 3.58, 95% CI 1.12–11.44) was significantly associated with an increased risk of fatigue. The OR values for anxiety and sleep disorders were 2.36 (95% CI 1.07–5.20) and 2.16 (95% CI 1.13–4.14) in females to males. The OR value of PTSD was 25.6 (95% CI 3.3–198.4) in patients with persistent symptoms to those without persistent symptoms. No significant associations were observed between fatigue syndrome or adverse mental outcomes and disease severity.

Conclusions

15-month follow-up in this study demonstrated the need of extended rehabilitation intervention for complete recovery in COVID-19 patients.