Characterization of tumour heterogeneity through segmentation-free representation learning on multiplexed imaging data
Related Articles
Segment Anything for Microscopy
Accurate segmentation of objects in microscopy images remains a bottleneck for many researchers despite the number of tools developed for this purpose. Here, we present Segment Anything for Microscopy (μSAM), a tool for segmentation and tracking in multidimensional microscopy data. It is based on Segment Anything, a vision foundation model for image segmentation. We extend it by fine-tuning generalist models for light and electron microscopy that clearly improve segmentation quality for a wide range of imaging conditions. We also implement interactive and automatic segmentation in a napari plugin that can speed up diverse segmentation tasks and provides a unified solution for microscopy annotation across different microscopy modalities. Our work constitutes the application of vision foundation models in microscopy, laying the groundwork for solving image analysis tasks in this domain with a small set of powerful deep learning models.
Systemic HER3 ligand-mimicking nanobioparticles enter the brain and reduce intracranial tumour growth
Crossing the blood–brain barrier (BBB) and reaching intracranial tumours is a clinical challenge for current targeted interventions including antibody-based therapies, contributing to poor patient outcomes. Increased cell surface density of human epidermal growth factor receptor 3 (HER3) is associated with a growing number of metastatic tumour types and is observed on tumour cells that acquire resistance to a growing number of clinical targeted therapies. Here we describe the evaluation of HER3-homing nanobiological particles (nanobioparticles (NBPs)) on such tumours in preclinical models and our discovery that systemic NBPs could be found in the brain even in the absence of such tumours. Our subsequent studies described here show that HER3 is prominently associated with both mouse and human brain endothelium and with extravasation of systemic NBPs in mice and in human-derived BBB chips in contrast to non-targeted agents. In mice, systemically delivered NBPs carrying tumoricidal agents reduced the growth of intracranial triple-negative breast cancer cells, which also express HER3, with improved therapeutic profile compared to current therapies and compared to agents using traditional BBB transport routes. As HER3 associates with a growing number of metastatic tumours, the NBPs described here may offer targeted efficacy especially when such tumours localize to the brain.
Identifying perturbations that boost T-cell infiltration into tumours via counterfactual learning of their spatial proteomic profiles
Cancer progression can be slowed down or halted via the activation of either endogenous or engineered T cells and their infiltration of the tumour microenvironment. Here we describe a deep-learning model that uses large-scale spatial proteomic profiles of tumours to generate minimal tumour perturbations that boost T-cell infiltration. The model integrates a counterfactual optimization strategy for the generation of the perturbations with the prediction of T-cell infiltration as a self-supervised machine learning problem. We applied the model to 368 samples of metastatic melanoma and colorectal cancer assayed using 40-plex imaging mass cytometry, and discovered cohort-dependent combinatorial perturbations (CXCL9, CXCL10, CCL22 and CCL18 for melanoma, and CXCR4, PD-1, PD-L1 and CYR61 for colorectal cancer) that support T-cell infiltration across patient cohorts, as confirmed via in vitro experiments. Leveraging counterfactual-based predictions of spatial omics data may aid the design of cancer therapeutics.
Cellpose3: one-click image restoration for improved cellular segmentation
Generalist methods for cellular segmentation have good out-of-the-box performance on a variety of image types; however, existing methods struggle for images that are degraded by noise, blurring or undersampling, all of which are common in microscopy. We focused the development of Cellpose3 on addressing these cases and here we demonstrate substantial out-of-the-box gains in segmentation and image quality for noisy, blurry and undersampled images. Unlike previous approaches that train models to restore pixel values, we trained Cellpose3 to output images that are well segmented by a generalist segmentation model, while maintaining perceptual similarity to the target images. Furthermore, we trained the restoration models on a large, varied collection of datasets, thus ensuring good generalization to user images. We provide these tools as ‘one-click’ buttons inside the graphical interface of Cellpose as well as in the Cellpose API.
A functional single-cell metabolic survey identifies Elovl1 as a target to enhance CD8+ T cell fitness in solid tumours
Reprogramming T cell metabolism can improve intratumoural fitness. By performing a CRISPR/Cas9 metabolic survey in CD8+ T cells, we identified 83 targets and we applied single-cell RNA sequencing to disclose transcriptome changes associated with each metabolic perturbation in the context of pancreatic cancer. This revealed elongation of very long-chain fatty acids protein 1 (Elovl1) as a metabolic target to sustain effector functions and memory phenotypes in CD8+ T cells. Accordingly, Elovl1 inactivation in adoptively transferred T cells combined with anti-PD-1 showed therapeutic efficacy in resistant pancreatic and melanoma tumours. The accumulation of saturated long-chain fatty acids in Elovl1-deficient T cells destabilized INSIG1, leading to SREBP2 activation, increased plasma membrane cholesterol and stronger T cell receptor signalling. Elovl1-deficient T cells increased mitochondrial fitness and fatty acid oxidation, thus withstanding the metabolic stress imposed by the tumour microenvironment. Finally, ELOVL1 in CD8+ T cells correlated with anti-PD-1 response in patients with melanoma. Altogether, Elovl1 targeting synergizes with anti-PD-1 to promote effective T cell responses.
Responses