The coming era of proteomics-driven precision medicine

      A stratification system for breast cancer based on basoluminal tumor cells and spatial tumor architecture

      Deep phenotyping of the health–disease continuum in the Human Phenotype Project

      Large-scale proteomics project seeks treatments for neurodegenerative diseases

      APOE ε4 carriers share immune-related proteomic changes across neurodegenerative diseases

      A pan-cancer plasma proteomic atlas reveals cancer biomarkers

      Lithocholic acid phenocopies the anti-ageing effects of calorie restriction

Proteomics is playing an increasingly prominent role in precision medicine. By studying the dynamics of protein expression and function at a systems level, it can reveal how the products of gene expression change across time and space. A team led by He Fuchu at the National Center for Protein Sciences published an article defining the core value of proteomics in precision medicine, reviewing progress in its technologies and applications, and setting out a framework for its development. The article notes that although genomics has succeeded in personalized medicine, research suggests that genetic factors may influence many diseases less than expected, so a deeper understanding of the proteome and its dynamics is needed to further improve treatment outcomes. The article details several key technologies, including protein expression profiling, post-translational modification analysis, protein–protein interaction network construction and spatial proteomics. Advances in these technologies—such as improvements in mass spectrometry, single-cell sample processing and large-scale data mining—now provide strong support for proteomics research. The article also mentions specific databases and resources that are essential to advancing the field. In addition, it describes breakthroughs in the human proteome reference map, proteome-driven innovation in health monitoring, the clinical translation of customized therapy, and the prospects and challenges of proteomics-driven "practice-based wisdom medicine." Proteomics is transforming how we understand, diagnose and treat disease, establishing its central position in precision medicine. Its progress opens new paths toward higher-level precision healthcare.

▼

Triple-negative breast cancer (TNBC) is an aggressive form of breast cancer with a poor prognosis, and neither its pathogenesis nor its treatment strategies are fully understood. In recent years, immune checkpoint inhibitors (ICIs) have brought new progress to TNBC treatment, but molecular markers that can predict treatment response remain very limited. Because TNBC is highly heterogeneous, with marked differences between patients in tumor cell phenotype and tumor microenvironment, precise stratification is especially important for prognostic assessment and treatment selection. Using imaging mass cytometry (IMC), the team of Bernd Bodenmiller at the University of Zurich performed single-cell spatial profiling of tumor samples from 215 TNBC patients, revealing 11 distinct tumor cell phenotypes that dominate in different patients. Among them, basal-luminal tumor cells (MC8) were significantly associated with poorer disease-free survival and were more common in lymph-node-positive patients. By further analyzing the spatial relationship between CD8⁺T cells and tumor cells, the team classified TNBC patients into five types: basal-luminal (BL), basal (B), luminal (L), non-basal/non-luminal (N) and inflamed (INF). This study establishes, at single-cell level, a direct link between TNBC phenotypic heterogeneity, spatial architecture and clinical outcome, and proposes a TNBC stratification system based on basal-luminal tumor cells and the spatial pattern of CD8⁺ T cells—offering new ideas and methods with important clinical significance for the management and treatment of TNBC.

▼

Although electronic health records (EHR) can generate clinical insights at scale, they lack the granularity needed to reveal the deeper biological mechanisms of health and disease. By contrast, longitudinal deep phenotyping, while more resource-intensive and involving fewer participants, can capture molecular and physiological dynamics in fine detail across multiple biological scales, thereby improving the discovery of novel biomarkers and disease pathways. The Human Phenotype Project (HPP) is a large-scale prospective longitudinal deep phenotyping study conducted by a global, multi-center team. It has enrolled approximately 28,000 participants, of whom more than 13,000 have completed their initial follow-up visit. The project aims to identify molecular markers with diagnostic, prognostic and therapeutic value, and to develop artificial intelligence (AI)-based disease prediction models. HPP collects a comprehensive range of data covering medical history, lifestyle, nutrition, anthropometry, blood tests, continuous glucose and sleep monitoring, imaging, and multi-omics (genome, transcriptome, microbiome, metabolome and immunome). The analysis revealed how phenotypes vary with age and ethnicity and, through comparison with healthy controls, delineated the molecular features of disease. The team also presented a multimodal foundation AI model trained by self-supervision on dietary and continuous glucose data, which outperformed existing methods in predicting disease onset. This study provides a new perspective on the role of multimodal data in disease development, underscores the importance of interdisciplinary collaboration, and lays a solid foundation for personalized medicine and research into causal mechanisms.

▼

As the global population ages, the incidence of neurodegenerative diseases (NDDs) such as dementia continues to rise. World Health Organization data show that dementia has become the seventh leading cause of death worldwide and the fourth among people over 70. Although monoclonal antibodies have recently been approved for the symptomatic treatment ofAlzheimer's disease, most NDDs still have no cure, imposing heavy economic, medical and emotional burdens. In response, the Global Neurodegenerative Proteomics Consortium (GNPC) was established, using high-throughput proteomics platforms to analyze more than 35,000 cross-sectional samples (serum, plasma and cerebrospinal fluid) from 23 NDD studies. By integrating these data with genome-wide association study (GWAS) results, transcriptomic data and gene ontology information, the researchers identified several key classes of proteins: those stable between healthy aging and NDD cases; those changing only with healthy aging; those altered as cognition progresses from normal to NDD; protein networks affected by both aging and NDD; and potential therapeutic targets as well as protein panels applicable to diagnosis and prognosis. GNPC provides an important resource for NDD research, accelerating understanding of the protein networks underlying disease heterogeneity and driving the discovery and translation of clinically valuable biomarkers.

▼

Recent research shows that the APOE ε4 gene may participate in the onset and progression of multiple neurodegenerative diseases (NDDs) by regulating immune function, acting as a shared driver. The Global Neurodegenerative Proteomics Consortium (GNPC) analyzed 11,270 cerebrospinal fluid and blood samples and, for the first time, systematically revealed the central role of APOE ε4 in Alzheimer's disease, Parkinson's disease, frontotemporal dementia and other NDDs. The results show that, whether or not disease is present, APOE ε4 carriers exhibit a consistent protein expression signature: enhanced inflammatory response and upregulated monocyte and T cell activity, with this immune signature independent of each disease's specific pathology (such as β-amyloid plaques in AD or α-synuclein deposition in PD). Validation in 262 human prefrontal cortex samples further confirmed the finding. The study also notes that although APOE ε4 can systematically alter protein composition, disease onset is still modulated by environmental factors such as hypertension and smoking, which explains why some carriers never develop disease in their lifetime. This work repositions APOE ε4 from a static "risk marker" to a dynamic "immune regulatory switch," offering new ideas for early identification and precise intervention across brain diseases. Blood tests based on this signature may be used for risk prediction, and interventions targeting the relevant immune pathways could become a new strategy for preventing NDDs.

▼

As cancer biology research has deepened, genomic and proteomic analyses have identified a variety of biomarkers. Traditional studies, however, have relied largely on invasive tissue biopsies, limiting clinical application. By contrast, liquid biopsies such as plasma proteomics, being non-invasive and easy to perform, show great potential. The team of Ding Chen at Fudan University performed mass spectrometry analysis on 2,251 plasma samples from China (including pre-treatment cancer patients, patients with benign proliferative disease, and healthy controls), charted a plasma proteome atlas, and identified proteomic subtypes through consensus clustering. The study compared proteomic changes before and after surgery to assess treatment effect, and built binary and multi-cancer classification models based on proteomic features and clinical indicators, validating their ability to distinguish cancer patients from healthy individuals in independent cohorts. The results reveal substantial heterogeneity across tumor types, with some proteomic subtypes strongly associated with specific cancers; changes in protein patterns after surgery can be used to monitor treatment response. In addition, combining proteomics with clinical indicators can distinguish pathological subtypes of certain tumors. This study provides strong evidence for the use of plasma proteomics in early cancer screening, subtyping and treatment monitoring, and lays an important foundation for precision diagnosis and treatment as well as new drug development.

▼

Aging is a complex process accompanied by the decline of cellular and tissue function, and it increases the risk of many chronic diseases. Calorie restriction (CR) has been shown to extend lifespan and improve health in a range of organisms, but it is difficult to maintain over the long term, and the key molecular mechanisms behind its metabolic benefits remain unclear. Finding compounds that can mimic the anti-ageing effects of CR is therefore of great significance. Using metabolomic analysis of serum from CR mice, the team of Lin Shengcai at Xiamen University found that AMPK was significantly activated and could reproduce the effects of CR in cultured cells. Using high-performance liquid chromatography–mass spectrometry, gas chromatography–mass spectrometry and capillary electrophoresis–mass spectrometry, the study identified 695 metabolites significantly altered by CR (341 polar metabolites and 354 lipids). Further experiments showed that lithocholic acid could mimic the lifespan-extending and anti-ageing effects of CR in nematodes, fruit flies and mice. Mechanistic studies showed that lithocholic acid acts through the TULP3–sirtuin–v-ATPase–AMPK axis and involves several potential target proteins. This work fills a gap in our understanding of how the body senses CR metabolic signals to delay aging, and provides both a theoretical basis and targets for developing new longevity drugs.

▼