Engineered CD47 protects T cells for enhanced antitumour immunity.

Adoptively transferred T cells and agents designed to block the CD47-SIRPα axis are promising cancer therapeutics that activate distinct arms of the immune system1,2. Here we administered anti-CD47 antibodies in combination with adoptively transferred T cells with the goal of enhancing antitumour efficacy but observed abrogated therapeutic benefit due to rapid macrophage-mediated clearance of T cells expressing chimeric antigen receptors (CARs) or engineered T cell receptors. Anti-CD47-antibody-mediated CAR T cell clearance was potent and rapid enough to serve as an effective safety switch. To overcome this challenge, we engineered the CD47 variant CD47(Q31P) (47E), which engages SIRPα and provides a 'don't eat me' signal that is not blocked by anti-CD47 antibodies. TCR or CAR T cells expressing 47E are resistant to clearance by macrophages after treatment with anti-CD47 antibodies, and mediate substantial, sustained macrophage recruitment to the tumour microenvironment. Although many of the recruited macrophages manifested an M2-like profile3, the combined therapy synergistically enhanced antitumour efficacy. Our study identifies macrophages as major regulators of T cell persistence and illustrates the fundamental challenge of combining T-cell-directed therapeutics with those designed to activate macrophages. It delivers a therapeutic approach that is capable of simultaneously harnessing the antitumour effects of T cells and macrophages, offering enhanced potency against solid tumours.

An engineered interleukin-11 decoy cytokine inhibits receptor signaling and proliferation in lung adenocarcinoma.

The cytokine interleukin (IL)-11 has been shown to play a role in promoting fibrosis and cancer, including lung adenocarcinoma, garnering interest as an attractive target for therapeutic intervention. We used combinatorial methods to engineer an IL-11 variant that binds with higher affinity to the IL-11 receptor and stimulates enhanced receptor-mediated cell signaling. Introduction of two additional point mutations ablates IL-11 ligand/receptor association with the gp130 coreceptor signaling complex, resulting in a high-affinity receptor antagonist. Unlike wild-type IL-11, this engineered variant potently blocks IL-11-mediated cell signaling and slows tumor growth in a mouse model of lung cancer. Our approach highlights a strategy where native ligands can be engineered and exploited to create potent receptor antagonists.

Whole-cell segmentation of tissue images with human-level performance using large-scale data annotation and deep learning.

A principal challenge in the analysis of tissue imaging data is cell segmentation-the task of identifying the precise boundary of every cell in an image. To address this problem we constructed TissueNet, a dataset for training segmentation models that contains more than 1 million manually labeled cells, an order of magnitude more than all previously published segmentation training datasets. We used TissueNet to train Mesmer, a deep-learning-enabled segmentation algorithm. We demonstrated that Mesmer is more accurate than previous methods, generalizes to the full diversity of tissue types and imaging platforms in TissueNet, and achieves human-level performance. Mesmer enabled the automated extraction of key cellular features, such as subcellular localization of protein signal, which was challenging with previous approaches. We then adapted Mesmer to harness cell lineage information in highly multiplexed datasets and used this enhanced version to quantify cell morphology changes during human gestation. All code, data and models are released as a community resource.

An engineered ligand trap inhibits leukemia inhibitory factor as pancreatic cancer treatment strategy.

Leukemia inhibitory factor (LIF), a cytokine secreted by stromal myofibroblasts and tumor cells, has recently been highlighted to promote tumor progression in pancreatic and other cancers through KRAS-driven cell signaling. We engineered a high affinity soluble human LIF receptor (LIFR) decoy that sequesters human LIF and inhibits its signaling as a therapeutic strategy. This engineered 'ligand trap', fused to an antibody Fc-domain, has ~50-fold increased affinity (~20 pM) and improved LIF inhibition compared to wild-type LIFR-Fc, potently blocks LIF-mediated effects in pancreatic cancer cells, and slows the growth of pancreatic cancer xenograft tumors. These results, and the lack of apparent toxicity observed in animal models, further highlights ligand traps as a promising therapeutic strategy for cancer treatment.