This paper details extensive research on how cancer-induced nerve injury (CINI) promotes resistance to anti-PD-1 immunotherapy. The study employed a variety of methods, including multiomics analysis of human patient cohorts with various cancers like cutaneous squamous cell carcinoma (cSCC), metastatic melanoma, and gastric cancer, alongside in vivo mouse models and in vitro co-culture assays involving neurons and cancer cells. Key findings suggest that cancer cells cause myelin degradation, leading to nerve injury, which triggers inflammatory pathways, specifically involving IFN-I and IL-6 signaling. This chronic neuroinflammation attracts immunosuppressive cells to the tumor microenvironment, ultimately dampening anti-tumor immunity and reducing the effectiveness of anti-PD-1 therapy, an effect that could be reversed by blocking the associated inflammatory pathways.

References:

• Baruch E N, Gleber-Netto F O, Nagarajan P, et al. Cancer-induced nerve injury promotes resistance to anti-PD-1 therapy[J]. Nature, 2025: 1-12.

This paper introduces MintFlow, a novel generative AI algorithm designed for mapping and reprogramming human tissue microenvironments using spatial transcriptomics data. MintFlow's core innovation is its ability to disentangle intrinsic cellular characteristics from gene expression changes induced by the local microenvironment, moving beyond purely descriptive spatial analysis. Applied to human diseases such as atopic dermatitis, cutaneous melanoma, and clear cell renal cell carcinoma (ccRCC), the model successfully identified spatially organized, disease-specific cell states and microenvironment-induced gene programs (MGPs). Crucially, MintFlow allows for in silico perturbations (virtual simulations) of tissue environments, predicting how cell removal or replacement could reprogram local cellular states and guiding the generation of translational therapeutic hypotheses.

References:

• Akbarnejad A, Steele L, Jafree D J, et al. Mapping and reprogramming human tissue microenvironments with MintFlow[J]. bioRxiv, 2025: 2025.06. 24.661094.

This paper details how macropinocytosis—a cellular process for scavenging extracellular fluids—is crucial for maintaining the identity of myofibroblasts (myCAFs), a subtype of cancer-associated fibroblasts (CAFs), in pancreatic cancer, specifically under glutamine stress. The authors, Zhang et al., demonstrate that inhibiting macropinocytosis in CAFs forces a shift from myCAFs to the inflammatory iCAF subtype via a MEK-ERK-dependent intermediate state. This CAF plasticity fundamentally alters the tumor microenvironment by reducing collagen deposition and increasing vascular expansion, which enhances the infiltration of T cells and improves the efficacy of both immunotherapy and chemotherapy in models of pancreatic ductal adenocarcinoma (PDAC). Consequently, the research positions macropinocytosis inhibition as a viable strategy for combination treatment to improve outcomes in PDAC.

References:

• Zhang Y, Ling L, Murad R, et al. Macropinocytosis maintains CAF subtype identity under metabolic stress in pancreatic cancer[J]. Cancer Cell, 2025, 43(9): 1677-1696. e15.

The article introduces TriPath, a deep-learning platform designed for the analysis of three-dimensional (3D) pathology samples to predict clinical outcomes, specifically patient prognostication for prostate cancer recurrence. Researchers argue that traditional two-dimensional (2D) pathology is limited by sampling bias and insufficient representation of intrinsically 3D tissue, while TriPath leverages volumetric imaging from modalities like open-top light-sheet microscopy (OTLS) and microcomputed tomography (microCT) to capture comprehensive morphology. Key findings indicate that 3D volume-based prognostication significantly outperforms 2D slice-based approaches and established pathologist baselines, demonstrating the clinical potential of AI-driven 3D pathology for better risk stratification. The platform uses a weakly supervised learning approach, requiring only patient-level outcome labels rather than extensive manual annotations. The study suggests that analyzing a larger tissue volume effectively mitigates sampling bias and accounts for tissue heterogeneity, improving prediction reliability.

References:

• Song A H, Williams M, Williamson D F K, et al. Analysis of 3D pathology samples using weakly supervised AI[J]. Cell, 2024, 187(10): 2502-2520. e17.

This paper details the introduction and validation of PathoPlex, a scalable and open-source framework for highly multiplexed imaging and analysis of formalin-fixed paraffin-embedded (FFPE) tissue specimens. This technology significantly advances spatial biology by enabling the detection of over 120 protein markers at subcellular resolution, exceeding the capabilities of existing commercial systems. PathoPlex combines iterative indirect immunofluorescence with a custom-built, low-cost 3D printing-based automation system for high-throughput sample processing and utilizes the accompanying spatiomic Python library for sophisticated pixel-level cluster analysis. The efficacy of PathoPlex is demonstrated through studies on mouse models of kidney disease and human clinical samples of diabetic kidney disease, revealing complex molecular and structural signatures associated with disease progression and therapeutic response. The framework aims to provide universal access to advanced multiplexed imaging, shifting the focus from solely cellular-level to integrative, pixel-based analysis of tissue pathology.

References:

• Kuehl M, Okabayashi Y, Wong M N, et al. Pathology-oriented multiplexing enables integrative disease mapping[J]. Nature, 2025: 1-11.

This article details research on a stratification system for breast cancer, specifically focusing on triple-negative breast cancer (TNBC). The study, led by Meyer et al., utilized imaging mass cytometry (IMC) to characterize tumor cell phenotypes and the spatial architecture of the tumor microenvironment (TME) in a large cohort of patients. Key findings include the identification of a poor-prognosis basoluminal tumor cell phenotype linked to rapid disease recurrence and the importance of tumor-CD8+ T cell interaction patterns for predicting good outcomes and immunotherapy response. The authors ultimately propose a five-subtype stratification system for breast cancer, integrating tumor cell cytokeratin expression and immune cell spatial dynamics, which they validate across multiple independent cohorts.

References:

• Meyer L, Jackson H W, Eling N, et al. A stratification system for breast cancer based on basoluminal tumor cells and spatial tumor architecture[J]. Cancer Cell, 2025, 43(9): 1637-1655. e9.

This review article focus on Circulating Tumor Cells (CTCs), which are cancer cells shed into the bloodstream. It offers an overview of CTCs, describing their molecular biology, importance in understanding cancer dissemination and metastasis, and their potential as a liquid biopsy tool. A significant portion of the review is dedicated to detailing the technological challenges and advances in isolating and characterizing these rare cells, contrasting CTC-based diagnostics with the more clinically common circulating tumor DNA (ctDNA). Finally, the article explores the current and future clinical applications of CTC analysis, including monitoring metastatic disease, detecting localized cancers, and informing therapeutic strategies like immunotherapy and targeted treatments.

References:

• Dai C S, Mishra A, Edd J, et al. Circulating tumor cells: Blood-based detection, molecular biology, and clinical applications[J]. Cancer Cell, 2025, 43(8): 1399-1422.

This article reports on a novel immunotherapy strategy for gastric and other gastrointestinal cancers using TFF2-MSA, a CXCR4 partial agonist peptide, in combination with anti-PD-1 treatment. Researchers demonstrate that TFF2-MSA synergistically enhances anti-PD-1 efficacy by selectively reducing immunosuppressive neutrophils (PMN-MDSCs), particularly the CXCR4-highly-expressing Hdc-GFP+ subset, in the tumor microenvironment and circulation. Mechanistically, TFF2-MSA acts systemically to suppress cancer-driven granulopoiesis in the bone marrow, contrasting with traditional CXCR4 antagonists which can exacerbate immature myeloid cell mobilization. The resulting immune landscape shows a robust activation of anti-tumoral CD8+ T cells, leading to significant tumor growth inhibition and increased survival in mouse models. Clinical relevance is supported by the finding that low TFF2 levels correlate with elevated PMN-MDSCs and poor prognosis in human gastric cancer patients.

References:

• Qian J, Ma C, Waterbury Q T, et al. A CXCR4 partial agonist improves immunotherapy by targeting immunosuppressive neutrophils and cancer-driven granulopoiesis[J]. Cancer Cell, 2025.

This article introduces Stereo-seq V2, an advanced spatial transcriptomics method designed for high-resolution total RNA mapping on formalin-fixed, paraffin-embedded (FFPE) tissue sections, which are commonly used in clinical settings. This updated technique utilizes random priming to overcome the limitations of traditional methods on degraded FFPE RNA, enabling the capture of non-coding RNAs (ncRNAs) and providing uniform gene body coverage. The source highlights the method's capability to profile host and pathogen transcriptomes simultaneously, as demonstrated in a Mycobacterium tuberculosis (Mtb)-infected mouse model and human samples, and its use in identifying tumor-associated alternative splicing (AS) events in triple-negative breast cancer (TNBC) tissues. Furthermore, Stereo-seq V2 is shown to be effective in elucidating spatial B cell receptor (BCR) clone dynamics, offering potential utility in biomedical research and personalized medicine.

References:

• Zhao Y, Li Y, He Y, et al. Stereo-seq V2: Spatial mapping of total RNA on FFPE sections with high resolution[J]. Cell, 2025.

This paper introduces PathoROB, the first systematic robustness benchmark for evaluating Foundation Models (FMs) used in digital pathology. The study investigates how susceptible these large-scale AI models are to learning non-biological technical features (like scanner hardware or staining variations) rather than focusing exclusively on biological signals. The authors present three novel metrics, including the robustness index, to quantify these deficits, demonstrating that limited robustness can lead to major diagnostic errors ("Clever Hans" learning) in downstream clinical tasks. Furthermore, the paper explores post-hoc robustification techniques like stain normalization and ComBat batch correction to mitigate these issues without requiring costly FM retraining, ultimately establishing that robustness evaluation is essential for safe clinical adoption and should be a core design principle for future pathology FM development.

References:

• Kömen J, de Jong E D, Hense J, et al. Towards Robust Foundation Models for Digital Pathology[J]. arXiv preprint arXiv:2507.17845, 2025.