Yu Wang’s research while affiliated with Chinese PLA General Hospital (301 Hospital) and other places

Peptide-drug-conjugates (PDCs) are emerging as promising alternatives to antibody-drug-conjugates (ADCs) in oncology. PDCs, with their smaller size (<10 kDa), can offer improved tumor penetration. Additionally, PDCs can potentially exhibit a better safety profile due to their shorter blood PK, absence of an Fc domain, and reduced immunogenicity. Furthermore, PDCs can be entirely chemically synthesized, providing a homogeneous chemical entity. Epidermal Growth Factor Receptor (EGFR) is a key target that is overexpressed in many tumor types. Current EGFR inhibitors primarily focus on mutant forms, creating a treatment gap for tumors with only wild-type overexpression. Carbonic Anhydrase IX (CAIX) represents another cancer-specific marker that supports tumor survival in hypoxic or acidic conditions and is frequently co-overexpressed with EGFR in cancers such as non-small cell lung cancer (NSCLC), head and neck cancers, and pancreatic cancers. SYNB011128, a bispecific PDC, exhibits high affinity and specificity for binding both EGFR and CAIX. Upon internalization, SYNB011128 releases MMAE, a potent tubulin inhibitor, through linker cleavage by tumor-specific proteases. In total, SYNB011128 employs three primary anti-cancer mechanisms: 1) Competing with EGF to inhibit the EGFR pathway. 2) Inhibiting CAIX enzyme function, crucial for cancer cell survival in hypoxic environments and resistance clone development. 3) Delivering the cytotoxic MMAE payload. Preclinical studies showed that SYNB011128 achieved complete tumor regression in multiple pancreatic and colon cancer models, outperforming EGFR-antibody-drug conjugates (EGFR-ADCs). Additionally, SYNB011128 showed efficacy in models of Osimertinib-resistant non-small cell lung cancer (NSCLC). With a molecular weight of 4 kDa, SYNB011128 exhibited rapid and deep tumor penetration. PDC induced biomarker was detected in both the periphery and core of the tumor within 6 hours post-dosing. Despite its relatively short blood half-life, SYNB011128 ensured sustained accumulation of the payload within tumors. This pharmacokinetic profile may reduce on-target toxicity and enhance safety. In mouse studies, SYNB011128 demonstrated no adverse effects, such as weight loss or liver/kidney damage, after 28 days of treatment at doses that induced tumor regression. Chemically, SYNB011128 is synthesized as a single enantiomer, without diastereomeric isomers. In summary, PDC SYNB011128 effectively targets both EGFR and CAIX, utilizing multiple anti-cancer mechanisms and achieving deep tumor penetration, which may lead to robust clinical efficacy. By binding to two cancer antigens to enhance tumor selectivity over normal tissues, and a pharmacokinetic profile of sustained tumor accumulation vs a short blood half-life, SYNB011128 may also have an improved safety margin. Citation Format Caihong Zhou, Michael Poss, Cheng Lu, Ruochi Zhang, Yu Wang, Yuanpeng Xiong, Xin Gao, Xiao Zhang, Yan Degenhardt. SYNB011128: A novel bispecific peptide-drug-conjugate targeting EGFR and CAIX [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1761.

The rapid accumulation of Electronic Health Records (EHRs) has transformed healthcare by providing valuable data that enhance clinical predictions and diagnoses. While conventional machine learning models have proven effective, they often lack robust representation learning and depend heavily on expert-crafted features. Although deep learning offers powerful solutions, it is often criticized for its lack of interpretability. To address these challenges, we propose DeepSelective, a novel end to end deep learning framework for predicting patient prognosis using EHR data, with a strong emphasis on enhancing model interpretability. DeepSelective combines data compression techniques with an innovative feature selection approach, integrating custom-designed modules that work together to improve both accuracy and interpretability. Our experiments demonstrate that DeepSelective not only enhances predictive accuracy but also significantly improves interpretability, making it a valuable tool for clinical decision-making. The source code is freely available at http://www.healthinformaticslab.org/supp/resources.php .

Subjective well-being (SWB) has recently gained significant attention as an important indicator of measuring quality of life. Based on the Twitter Sentiment Geographical Index, this study analyzes the spatial distribution characteristic of county-level SWB across the United States. We apply multiple linear regression, structural equation modeling (SEM), and geographically weighted SEM to identify the overall effects of urban sprawl on SWB, examine their direct and indirect pathways, and investigate the spatial variations in these pathways. Our results show that SWB is spatially heterogeneous across U.S. counties, with high levels of SWB in inland U.S. counties and low levels of SWB in U.S. counties along the coast, near the Great Lakes, and bordering other countries. Both population and employment sprawl can generally improve SWB. Although population and employment sprawl can directly enhance SWB, they also have contrasting indirect effects: Population sprawl tends to reduce SWB by limiting upward mobility opportunities and potentially increasing violent crime, whereas employment sprawl often mitigates these issues. Additionally, population sprawl increases SWB by mitigating air pollution. Notably, both direct and indirect pathways exhibit significant spatial differences. These findings contribute to a nuanced perspective on the intricate relationships between urban sprawl and SWB, highlighting the need to consider spatial variations.

Health inequity represents a significant social injustice with major policy implications. This study examines the role of neighborhood intergenerational mobility (IM)—defined as the extent to which children within a specific neighborhood can achieve better socioeconomic outcomes than their parents—in shaping population health, addressing widening health inequalities. We propose that neighborhood IM is positively associated with population health, moderated by spatial dependency and heterogeneity. Analyzing over 69,000 census tracts in the contiguous United States using spatially-lagged X models, we find that neighborhood IM is positively associated with health status. The positive relationship weakens in neighborhoods surrounded by neighborhoods with higher levels of IM and strengthens in neighborhoods surrounded by neighborhoods with lower levels of IM. It also weakens in more advantaged environments—characterized by higher socioeconomic indicators, better built environment features, and more favorable natural environment conditions—and strengthens in less advantaged environments with poorer socioeconomic, built, and natural conditions. Our findings underscore the critical role of neighborhood context and heterogeneity in shaping the effects of social determinants on health, suggesting that policymakers should prioritize resources for disadvantaged neighborhoods with lower IM, particularly those surrounded by similarly low-IM areas, to mitigate health disparities more effectively. Our study provides new insights into the role of neighborhood IM in population health and demonstrates the value of geographic approaches for understanding and mitigating health disparities.

Health inequity represents a significant social injustice with major policy implications. This study examines the role of neighborhood intergenerational mobility (IM)—defined as the extent to which children within a specific neighborhood can achieve better socioeconomic outcomes than their parents—in shaping population health, addressing widening health inequalities. We propose that neighborhood IM is positively associated with population health, moderated by spatial dependency and heterogeneity. Analyzing over 69,000 census tracts in the contiguous United States using spatially-lagged X models, we find that neighborhood IM is positively associated with health status. The positive relationship weakens in neighborhoods surrounded by neighborhoods with higher levels of IM and strengthens in neighborhoods surrounded by neighborhoods with lower levels of IM. It also weakens in more advantaged environments—characterized by higher socioeconomic indicators, better built environment features, and more favorable natural environment conditions—and strengthens in less advantaged environments with poorer socioeconomic, built, and natural conditions. Our findings underscore the critical role of neighborhood context and heterogeneity in shaping the effects of social determinants on health, suggesting that policymakers should prioritize resources for disadvantaged neighborhoods with lower IM, particularly those surrounded by similarly low-IM areas, to mitigate health disparities more effectively. Our study provides new insights into the role of neighborhood IM in population health and demonstrates the value of geographic approaches for understanding and mitigating health disparities.

Objective: To investigate the impact of Meniere's Disease (MD) on balance and proprioception by utilising multi-frequency Vestibular Evoked Myogenic Potentials (VEMP) to evaluate otolith function. Design: Observational study employing the Otolith Tuning Index (OTI) to quantify vestibular function through analysis of VEMP response rates and tuning ratios. Study sample: A total of 123 participants were included, comprising 94 patients diagnosed with MD and 29 healthy controls. VEMP testing was conducted at frequencies of 500 Hz, 750 Hz, and 1 kHz. Results: Among MD patients, 69% reported imbalance, with severe cases predominating in advanced stages. The non-response rate for oVEMP at 500 Hz was 73.3% on the affected side, associated with unpredictable falls. Significant correlations were observed between cVEMP non-responses and both disease severity (p = 0.012) and walking imbalance (p = 0.037). oVEMP responses were lowest at 500 Hz, improving at 1 kHz, whereas cVEMP amplitudes peaked at 500 Hz bilaterally. OTI values indicated significant otolith dysfunction on affected sides compared to contralateral sides and controls (p = 0.026, p = 0.032, p < 0.001), with dysfunction worsening with disease progression and age. Conclusions: The Otolith Tuning Index (OTI) effectively measures otolith dysfunction in MD patients, offering valuable insights to enhance diagnostics, patient management, and treatment planning.

The development of highly integrated electronic components and the Internet of Things demands efficient thermal management and uninterrupted energy harvesting, which provides exciting opportunities for thermoelectric (TE) technology since it allows direct conversion between electricity and thermal energy. The improved output performance of TE devices has traditionally been driven by advancements in inorganic materials. Recently, there has been growing interest in studying substrate‐free inorganic‐based TE thin films because they provide improved adherence to curved surfaces and offer a more compact size compared to the corresponding rigid form of these materials. This review begins by summarizing various methods for fabricating freestanding inorganic‐based TE films, including leveraging the intrinsic plasticity of certain materials, exfoliating layered‐structure materials, using sacrificial substrates, and creating composites with flexible components such as polymers and carbon‐based materials. A key challenge in achieving high device performance is determining how to maintain the favorable TE properties of inorganic materials. This can be addressed through strategies such as high inorganic content loading, multicomponent engineering, and interfacial structure design. The review also discusses the applications of substrate‐free inorganic‐based TE devices in both power generation and solid‐state cooling. Finally, it outlines current challenges and proposes potential research directions to further advance the field.