Theme 1: Promoting Good Health
Project Title: Multi-disciplinary Approaches to Tackle the Global Public Health Threat of Hypervirulent and Multidrug-resistant Klebsiella Pneumoniae
Project Coordinator: Prof Sheng Chen (CityU)
Abstract
The Gram-negative bacterial pathogen Klebsiella pneumoniae has consistently evolved over the past two decades, generating genetic variants of mixed phenotypes of antibiotic resistance and virulence. Among them, the newly emerged multidrug resistant (MDR) and hypervirulent strains are of particular concern. To date, K. pneumoniae has already become the most frequently isolated bacterial pathogen in hospital settings and the most common pathogen that causes blood stream infections, with mortality rate of over 40% recorded in various countries. However, the threat imposed by new variants of this old pathogen has not been recognized. A thorough understanding of the on-going evolution trend, transmission dynamics and pathogenic mechanisms of this notorious pathogen is essential for development of effective intervention strategies to prevent occurrence of a hypervirulent and MDR K. pneumoniae pandemic. To this end, we propose to adopt comprehensive, multidisciplinary study approaches to depict the genetic basis of the varied phenotypic features of these organisms and identify molecular markers for design of effective methods to differentiate between high and low risk strains, and develop novel therapies to treat MDR and hypervirulent K. pneumoniae infections. As an internationally recognized team in K. pneumoniae research supported by the HMRF, GRF, CRF and RIF grants, we conducted the first national surveillance of carbapenem-resistant K. pneumoniae (CRKP) in China, identified the first cases of ST23 type hypervirulent and carbapenem-resistant K. pneumoniae, discovered the first ST11 and hypervirulent and carbapenem-resistant K. pneumoniae strain, and recently reported the first conjugative virulence plasmid that augmented the virulence level of K. pneumoniae. We have already established a collection of 7 000 clinical strains and identified novel drug candidates that exhibit high efficacy, both in vitro and in vivo, in eradicating both multidrug and hypervirulent variants of K. pneumoniae. Support by the Theme-based Research Grant is necessary for implementation of this strategic research plan to devise feasible pre-emptive approaches to halt global dissemination of key pathogenic K. pneumoniae variants and protect human health.
Theme 1: Promoting Good Health
Project Title: Delineating and Translating the Mechanistic Determinants to Improve the Clinical Management of Liver Cancer
Project Coordinator: Prof Irene Oi-Lin Ng (HKU)
Abstract
Liver cancer (Hepatocellular carcinoma, HCC) is one of the commonest malignancies worldwide and highly prevalent in this region. It is an aggressive cancer and often diagnosed at advanced stages and hence not operable. Immunotherapeutics and targeted therapy are given to patients with advanced tumors, but often on a 'one-size-fits-all' basis. Mechanistic determinants and biomarkers are much needed to guide treatment to improve patient outcome. To address this issue, we will use a multi-pronged approach consisting of three distinct but highly interconnected programs. First, while we will establish the efficacy of immune checkpoint inhibitor (ICI) treatment in giving better surgical resectability, we will delineate and spatially reconstruct the genomic, molecular and cellular landscapes in determining the underlying mechanisms for immunotherapy response, incorporating state-of-the-art technologies (various sequencing technologies including spatial transcriptomics). Second, preclinical evaluation, functional validation, and mechanistic characterization will be performed using in vivo modelling with genome-editing mouse models to precisely reproduce different genetic backgrounds especially those identified in Objective 1. This will enable strategic preclinical testing and dissecting the mechanism of oncogenic mutations and epigenetic alterations in shaping ICI treatment response. Third, blood-based biomarkers devised from cell-free DNA signatures using tailor-made targeted-panel and from targeted exosome profiling will be established to inform treatment response. We aim to translate the biomarkers at different clinical disease stages and prospective multi-timepoint follow-up to inform treatment efficacy and tracking disease progression and recurrence. Furthermore, proteolysis targeting chimera (PROTAC) platform will be adopted to degrade specific protein targets in HCC to find effective targeted therapy. Our track records - The team has long-standing and vast experience on HCC: liver cancer stemness, cancer stem cells, genomic profiling, clinical management, new Hong Kong HCC staging system, and clinical trials. We have also established various cutting-edge technology platforms such as single-cell transcriptomics, genome-editing mouse models, genome-wide CRISPR / Cas9 knockout models, and comprehensive functional characterization of gene targets, and exosomes. Clinical samples - We have collected huge, well-annotated banks consisting of thousands of blood and frozen tissue samples, and established patient-derived HCC tumor xenograft lines. These expertise, cutting-edge technologies and resource have put us on the international map of liver cancer research. They also form the solid grounds for this proposed program. Impact - This program will generate unique information for evidence-based translational application to improve diagnosis and treatment outcome for patients with this deadly cancer. This will save patients' health, and financial expenses, and cut social health care expenses. It will put Hong Kong as an international hub for HCC.
Theme 1: Promoting Good Health
Project Title: Characterization of Tumor Microenvironment in Nasopharyngeal Carcinoma
Project Coordinator: Prof Xin-Yuan Guan (HKU)
Abstract
Nasopharyngeal carcinoma (NPC) is a cancer of strategic importance in Hong Kong due to its high incidence among Southern Chinese. According to our latest randomized clinical trials, >30% of the most advanced locoregional NPC patients relapse despite the best chemo-radiotherapy available. NPC recurrence and treatment resistance remain major challenges in the clinical management of NPC. For patients with recurrence of metastatic diseases, combination of immunotherapy and chemotherapy treatment has emerged as a promising strategy. However, despite the heavy infiltration of lymphocytes in tumor tissues, clinical experience showed that only 20-30% of NPC patients respond to PD-1 inhibitors. In addition, none of currently used target therapies showed impressive clinical efficacy for NPC due to lack of reliable preclinical models for drug evaluation and patient stratification. Hence, there is an unmet clinical need for developing patient-specific strategy to enhance treatment response against advanced and metastatic NPC based on a better understanding of TME. In Aim 1 of the proposal, multiomics, computational and mechanistic approaches will be deployed to elucidate the microenvironmental landscape of NPC in a single-cell spatiotemporal resolution. Several technical platforms, such as organoid, humanized mouse model and multiplex immunohistochemistry will be established to characterize the TME of NPC. In Aim 2, important findings revealed by single cell sequencing, including CD70-mediated immune suppression, metabolic checkpoints in the NPC-infiltrated T cells and its influences on exhaustion and immunosuppression will be further investigated. Data generated by single cell sequencing will be also applied to study the cell-cell interactions and communication within TME in NPC by bioinformatics. In addition, genome-wide CRISPR/Cas9 knockout/repression screenings will be used to identify more genes and pathways involved in PDL1 regulation in NPC. The main task in Aim 3 is to develop novel NPC therapeutic interventions, including establishing predictive models for evaluation of immunotherapeutic response in NPC patients, establishing a Deep-Learning model incorporating multiple features for predicting response to PD-1/PDL-1 checkpoint inhibitors in NPC, investigating peripheral immunological response to treatment with checkpoint inhibitor in recurrent/metastatic NPC and testing combination treatment of anti-PD-1/anti-PD-L1 with chemotherapy and target therapy in NPC. Our proposed research will lead to quantum leap in advancing personalized strategy and precision therapy in treatment based on accurate prediction of cancer progression and therapeutic responsiveness, as well as specific targets in individual NPC patient. These will ultimately result in major breakthroughs for saving more lives with greater efficacy, less toxicity, and lower cost.
Theme 2: Developing a Sustainable Environment
Project Title: Towards Carbon Neutrality: Catalysing H2O and CO2 to Green Resource Carriers
Project Coordinator: Prof Zhengxiao Guo (HKU)
Abstract
Rapid transition towards net-zero carbon (CO2) emission is an imperative and complex undertaking by science and society to stave off catastrophic climate change, while sustaining resource securities in energy, food, water and materials. If there were a panacea, it would be a "water-carbon (H2-CO2) economy", where resources can be effectively inter-converted and/or resourcified, i.e. to manage a few chemical bonds among H-C-O(-N) atoms. Nature (bio-) does it better but is "huffing & puffing" with exhaustion and too slow. Hence, this project aims to enable a multi-disciplinary "Green Resource Carriers Hub" in Hong Kong to catapult effective mechanistic and technological developments of green resource carriers, starting with hydrogen and methanol, along the most effective techno-economic pathways. Renewable solar/wind and off-peak electrochemical conversion will drive the conversion of H2O (potentially seawater) and greenhouse gas CO2, respectively day and night, to enhance both efficiency and implementation. We will combine in-silico deep-learning and simulation design with experimental engineering to realize the goals through five interactive Work-Packages (WPs). New catalytic nanostructures and active microenvironments for H2O and CO2 electrolyzers and photo-reactors will be created (WP1,2), and their performance will be characterized using both ex-situ and in-situ techniques under operando conditions (WP3). These know-hows will be integrated into a prototype to demonstrate the practicality of the new technology (WP4). Socio-techno-economic analysis of the green resource carriers will be incorporated with a specific focus on their future influence and impact towards HK's 2050 carbon neutrality goal (WP5.1). The Management will oversee the progress of all parties in the consortium, the comprehensive dissemination and knowledge exchange processes (WP5.2). The project will deliver on three key a-/re-venues: 1) in-depth knowledge of effective structural features on H2 and CO2 conversion pathways, their enhancement by electric- / photo- field potential, and selectivity of final products; 2) integrated methodologies and effective catalysts / photo-catalysts for green resource carriers; and 3) demonstrator devices (H2O and CO2 electrolyzers and photo-reactors) for flexible deployment of the technologies. An Advisory Board will oversee the overall progress to ensure high-quality deliverables in scientific knowledge, technologies and socio-techno-economic pathways (STEPs) for their potential impact in HK.
Theme 2: Developing a Sustainable Environment
Project Title: Unravelling the Black Box between Air Pollution and Public Health for Transformative Air Quality Management
Project Coordinator: Prof Xiang-dong Li (PolyU)
Abstract
Air pollution is the greatest environmental health risk factor for premature deaths worldwide; among the air pollutants, fine particulate matter (PM2.5) is one of the biggest concerns. PM2.5 can penetrate deep into and accumulate in the human lungs and other body parts. Every year, ambient PM2.5 pollution causes millions of premature deaths and costs the global economy US$225 billion of labour income losses. Many regulatory bodies, including the Hong Kong Environmental Protection Department (EPD), benchmark against the guideline values issued by the World Health Organization (WHO) (e.g., mass concentration of PM2.5, µg m-3) to protect public health. However, evidence is mounting on the differing health effects of equal mass concentrations of PM2.5 samples collected over a wide range of time and from different locations. This is primarily because PM2.5 is a cocktail of components from a mix of sources. Not all components or sources are equally important in terms of toxicity contribution when assessing their combined impact on air quality. Controlling the sources of health-relevant fractions of PM2.5 instead of the mass-dominating ones would be a more effective option than managing the entire PM2.5 mass. Identifying the toxic components and their associated sources responsible for PM2.5 health effects represents a major scientific challenge prior to policy-making. The recent advances in environmental toxicology and molecular epidemiology provide opportunities towards solving the long-standing puzzle. Supported by an existing PM2.5 global monitoring network, we will select multiple cities representative of distinct natural and socioeconomic conditions in the study. We will decipher the responsible PM2.5 components and emission sources underlying the increased exacerbation risks of two index diseases, chronic obstructive pulmonary disease (COPD) and ischemic heart disease (IHD). We will evaluate the benefits versus costs of our proposed strategy of targeting the sources of PM2.5 toxic constituents against the conventional approach targeting total mass concentrations. Effective and economical approaches to manage air quality and public health will be recommended for city- or country-specific scenarios.
Theme 3: Enhancing Hong Kong's Strategic Position as a Regional and International Business Centre
Project Title: SynchronHub: Cyber-Physical Internet for Synchronizing Cross-Border Logistics Hubs in the Greater Bay Area (GBA)
Project Coordinator: Prof George Q. Huang (HKU)
Abstract
The vision of Physical Internet (PI) is to send and receive goods just like sending and receiving email messages. Europe has made great efforts in this field. If realized, the way that logistics services are provided and consumed will change dramatically, just as email changed the role of post offices and the way people use postal services. The Greater Bay Area (GBA) has an excellent transportation infrastructure. Hong Kong, Guangzhou and Shenzhen are three premier land, marine, aviation and rail hubs through which cities are well connected. This motivates us to explore and build PI in the GBA. This project proposes to innovate four pillars for Cyber-Physical Internet (CPI): (1) CPI digitization technologies for creating cyber-physical logistics systems; (2) CPI network services for setting, configuring and operating plug-and-play components; (3) CPI mechanisms to motivate and facilitate collaboration; and (4) CPI decision supports for synchronized logistics planning, scheduling and execution. The vision of CPI is to send and receive goods just like sending and receiving messages within chat groups using instant messaging platforms. The GBA economy has been dominated by export manufacturing activities. Import of raw materials and export of finished products both start and end in overseas markets, and both ends are well connected through cost-effective logistics hubs of multiple modes. However, global manufacturing has recently experienced substantial reconfiguration under reindustrialization strategies of many developed economies, including Hong Kong. China's National 14th 5-Year Plan calls for a "dual circulation" development pattern. The GBA has to transform from once the "Factory of the World" to meeting both domestic and global demands. The transformation has been unprecedentedly urged by the Covid-19 pandemic. Social distancing measures at global and local levels have resulted in large-scale interruptions in logistics operations. People are forced to work from home and use online shopping. Demands for high-quality e-commerce logistics services have soared. But, interrupted port operations, cancelled passenger flights, and delayed shipment arrivals and departures have substantially constrained capacities and created serious operational jams and deadlocks at terminals and ports. The world container index has increased over five times and the airfreight index more than doubled in two years. CPI contributes to establishing post-pandemic "new norms", while logistics resilience and CO2 emission targets are achieved. We have gathered a multidisciplinary team of world-leading researchers with complementary expertise to find answers to the fundamental research questions and technological challenges required to innovate CPI solutions. This project is supported by logistics business associations and leading companies in Hong Kong, GBA cities and global partners.
Theme 4: Advancing Emerging Research and Innovations Important to Hong Kong
Project Title: Institute of Medical Intelligence and XR
Project Coordinator: Prof Pheng Ann Heng (CUHK)
Abstract
Technological innovation presented new and promising ways to improve medical diagnosis, treatment, education, and healthcare service with ever-increasing rigor, subtlety, insight, and precision. Among these advanced technologies, artificial intelligence (AI) and extended reality (XR) are growing fast and making huge transformations in medicine and healthcare. XR refers to all real-and-virtual combined environments and human-machine interactions generated by computer technology and wearables. Integrating AI and XR can open many possibilities towards delivering precision medicine for next-generation healthcare. Yet, there remain challenges to apply AI / XR to medical image computing and computer-assisted intervention in real-world clinical applications. Our project's objective is to build a world-class institute for medical intelligence and XR in Hong Kong by developing cutting-edge techniques aimed at overcoming these challenges and facilitating “one-stop” medicine and healthcare services, covering screening, diagnosis, treatment, management, and prognosis. Specifically, we will address the following major challenges and questions: (1) How does medical intelligence help precision medicine? (2) How can intelligent data analytics be delivered to clinicians/patients in a human-centered way with AI and XR? (3) How can intuitive AI-enabled interaction be facilitated for clinicians with future intelligent XR systems? We will innovate solutions to address these challenges: (1) Intelligent Personalized Diagnosis (IPD) for diagnosis via advanced medical image analysis; (2) AI-XR Interaction & Virtual Surgery (IVS) for next-generation visualization, assessment, treatment coordination, precise planning, and surgical training and education; (3) Intraoperative AI-AR Assisted Surgery (IAS) for dynamic medical interpretation, efficient execution of a surgical plan, and real-time fused intraoperative image guidance; and (4) an integrated multi-facet pipeline platform for applications to liver cancer (hepatocellular carcinoma) and kidney cancer diagnosis, treatment, prognosis, and medical training. Our research team has collaborated fruitfully over a long period. Our backgrounds and strengths are complementary, enabling synergies to be achieved. Ultimately, our collective efforts will advance the frontier of AI and XR in medical and healthcare applications.
Theme 4: Advancing Emerging Research and Innovations Important to Hong Kong
Project Title: ReRACE: ReRAM AI Chips on the Edge
Project Coordinator: Asso Prof Ngai Wong (HKU)
Abstract
The unprecedented development in machine learning (ML) and artificial intelligence (AI) in the past decade has spawned a new era of smart living. A recent study on AI hardware by McKinsey highlights that among central processing units (CPUs), graphics-processing units (GPUs), field-programmable gate arrays (FPGAs) and application-specific integrated circuits (ASICs, aka chips), AI on ASICs is forecast to account for 50% to 70% of all AI solutions in 2025. This is echoed by Yahoo Finance predicting a growth of $73.49 billion in the AI chip market from 2021 to 2025. At the heart of AI is deep learning and deep neural networks (DNNs) which have posed new challenges due to their ever-growing complexity and sizes. This contrasts with the latest AI trend to offload various AI tasks to the edge (namely, terminal or user-end) equipped with only resource-constrained hardware. The situation is accentuated by the traditional von Neumann computing architecture which suffers power and latency bottlenecks due to the heavy data traffic between memory and processing elements.
To sustain the evolution of state-of-the-art neural architectures and to enable low-power, high-speed edge AI, we need to shift away from the quantitative scaling in modern transistor technologies and explore a qualitative transform of microelectronics platform. Subsequently, this project sets forth to realize in-memory computing by utilizing the next-generation microelectronic memristor devices, among which the resistive random-access memory (ReRAM) stands out due to its low power, high reliability and multi-level programmability that are perfect for edge AI.
This project is codenamed ReRACE (ReRAM AI Chips on the Edge) and pulls together a strong team to tackle various technical challenges. Specifically, it aims to achieve a complete device-circuit and system-application stack that harnesses ReRAM arrays for ultra-low-power and high-speed edge AI computing. Outputs from ReRACE will provide a versatile solution for next-generation edge AI chips featuring in-memory computing and orders of improvement in energy efficiency over existing von Neumann computing architectures. The success of ReRACE is warranted by its team formation comprising world-class domain experts encompassing the ReRAM device-circuit and system-application layers, and further backed by the support from industry and the latest advanced microelectronics infrastructure in Hong Kong. To sum up, ReRACE is extremely timely and carries high research and practical impacts. Its deployment can readily catalyse Hong Kong into a regional hub of next-generation edge AI chips, foster talents for the long-term microelectronics R&D in Hong Kong, and contribute to her strategic goal of high-tech reindustrialization.