SignificanceIpilimumab, an antibody that recognizes cytotoxic T lymphocyte antigen (CTLA)-4, was the first approved “checkpoint”-blocking anticancer therapy. In mice, the response to antibodies against CTLA-4 depends entirely on expression of the Fcγ receptor. We developed H11, an alpaca heavy chain-only antibody fragment against CTLA-4 that lacks an Fc portion and inhibits interactions between CTLA-4 and its ligand. By using H11 to visualize CTLA-4 expression in the whole animal, we found that accessible CTLA-4 is largely confined to the tumor; however, H11 treatment has minimal effects on antitumor responses. Installing the murine IgG2a constant region on H11 greatly enhances antitumor response. We were thus able to dissociate CTLA-4 blockade from CTLA-4–dependent receptor engagement as an explanation for the antitumor effect.

Programmed death ligand 1 (PD-L1) is expressed on a number of immune and cancer cells, where it can downregulate antitumor immune responses. Its expression has been linked to metabolic changes in these cells. Here we develop a radiolabeled camelid single-domain antibody (anti-PD-L1 VHH) to track PD-L1 expression by immuno-positron emission tomography (PET). PET-CT imaging shows a robust and specific PD-L1 signal in brown adipose tissue (BAT). We confirm expression of PD-L1 on brown adipocytes and demonstrate that signal intensity does not change in response to cold exposure or β-adrenergic activation. This is the first robust method of visualizing murine brown fat independent of its activation state.

CD47 is an antiphagocytic ligand broadly expressed on normal and malignant tissues that delivers an inhibitory signal through the receptor signal regulatory protein alpha (SIRPα). Inhibitors of the CD47-SIRPα interaction improve antitumor antibody responses by enhancing antibody-dependent cellular phagocytosis (ADCP) in xenograft models. Endogenous expression of CD47 on a variety of cell types, including erythrocytes, creates a formidable antigen sink that may limit the efficacy of CD47-targeting therapies. We generated a nanobody, A4, that blocks the CD47-SIRPα interaction. A4 synergizes with anti-PD-L1, but not anti-CTLA4, therapy in the syngeneic B16F10 melanoma model. Neither increased dosing nor half-life extension by fusion of A4 to IgG2a Fc (A4Fc) overcame the issue of an antigen sink or, in the case of A4Fc, systemic toxicity. Generation of a B16F10 cell line that secretes the A4 nanobody showed that an enhanced response to several immune therapies requires near-complete blockade of CD47 in the tumor microenvironment. Thus, strategies to localize CD47 blockade to tumors may be particularly valuable for immune therapy.

The dense network of interconnected cellular signalling responses that are quantifiable in peripheral immune cells provides a wealth of actionable immunological insights. Although high-throughput single-cell profiling techniques, including polychromatic flow and mass cytometry, have matured to a point that enables detailed immune profiling of patients in numerous clinical settings, the limited cohort size and high dimensionality of data increase the possibility of false-positive discoveries and model overfitting. We introduce a generalizable machine learning platform, the immunological Elastic-Net (iEN), which incorporates immunological knowledge directly into the predictive models. Importantly, the algorithm maintains the exploratory nature of the high-dimensional dataset, allowing for the inclusion of immune features with strong predictive capabilities even if not consistent with prior knowledge. In three independent studies our method demonstrates improved predictions for clinically relevant outcomes from mass cytometry data generated from whole blood, as well as a large simulated dataset. The iEN is available under an open-source licence.

This diagnostic/prognostic study describes the use of cell-free transcriptomics, urine metabolomics, and plasma proteomics for identifying the biological measurements associated with preterm birth.

High-throughput single-cell analysis technologies produce an abundance of data that is critical for profiling the heterogeneity of cellular systems. We introduce VoPo (https://github. com/stanleyn/VoPo), a machine learning algorithm for predictive modeling and comprehensive visualization of the heterogeneity captured in large single-cell datasets. In three mass cytometry datasets, with the largest measuring hundreds of millions of cells over hundreds of samples, VoPo defines phenotypically and functionally homogeneous cell populations. VoPo further outperforms state-of-the-art machine learning algorithms in classification tasks, and identified immune-correlates of clinically-relevant parameters.

A multitude of clinical, biological, environmental, and demographic factors influence the trajectory of a pregnancy. Maternal genetics, environment, stress, nutrition, medical history, socioeconomic status, and racial and ethnic background all play a role in determining the success of a pregnancy. Diverse data sources are available for the study of pregnancy and prediction of adverse outcomes, including electronic health records (EHRs) and administrative claims data, high-throughput multiomics data for characterizing biological systems, and more complex sources like time series, imaging and video data, and text. Recent advances in multiview, multitask, and deep learning allow joint modeling across data sources as well as across outcomes and demonstrate the vast potential of such integrated approaches.

No abstract