21 May 2018 in Journal of Medical Internet Research
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet
3 May 2024 in Science Immunology
Understanding the mechanisms that regulate T cell immunity is critical for the development of effective therapies for diseases associated with T cell dysfunction, including autoimmune diseases, chronic infections, and cancer. Co-inhibitory “checkpoint molecules,” such as programmed cell death protein-1, balance excessive or prolonged immune activation by T cell–intrinsic signaling. Here, by screening for mediators of natural killer (NK) cell recognition on T cells, we identified the immunoglobulin superfamily ligand B7H6 to be highly expressed by activated T cells, including patient-infused CD19-targeting chimeric antigen receptor (CAR) T cells. Unlike other checkpoint molecules, B7H6 mediated NKp30-dependent recognition and subsequent cytolysis of activated T cells by NK cells. B7H6 + T cells were prevalent in the tissue and blood of several diseases, and their abundance in tumor tissue positively correlated with clinical response in a cohort of patients with immune checkpoint inhibitor–treated esophageal cancer. In humanized mouse models, NK cell surveillance via B7H6 limited the persistence and antitumor activity of CAR T cells, and its genetic deletion enhanced T cell proliferation and persistence. Together, we provide evidence of B7H6 protein expression by activated T cells and suggest the B7H6-NKp30 axis as a therapeutically actionable NK cell–dependent immune checkpoint that regulates human T cell function.
28 February 2025 in Science Immunology
The bone marrow microenvironment plays a crucial role in the development of multiple myeloma. As the disease progresses, malignant myeloma cells can evolve to survive outside the bone marrow. However, the processes underlying bone marrow independence and their consequences for immune control remain poorly understood. Here, we conducted single-cell and spatial multiomics analyses of bone marrow–confined intramedullary disease and paired breakout lesions that disrupt the cortical bone. These analyses revealed a distinct cellular microenvironment and architectural features of breakout lesions, characterized by extensive areas of malignant plasma cells interspersed with lesion-specific solitary natural killer and macrophage populations, as well as focal accumulations of immune cell agglomerates. Within these agglomerates, spatially confined T cell clones expanded alongside various immune cells, coinciding with the local genomic evolution of tumor cells. These analyses identify breakout lesions as a hotspot for tumor-immune cell interactions and diversification, representing a key event in myeloma pathogenesis.
30 June 2026 in Nature Medicine
Multidisciplinary tumor boards integrate longitudinal treatment histories, molecular profiling and rapidly evolving evidence to guide decisions in hematological malignancies, yet access to this level of subspecialty deliberation is increasingly uneven. Here we develop HemaGuide, a locally deployable, modular large language model agent that converts unstructured clinical documents into structured case representations, autonomously routes cases to specialized decision modes (‘guideline’, ‘advanced’ and ‘molecular’) and grounds recommendations in disease-specific guideline flowcharts and a clinical decision memory of >2,000 real-world tumor board cases. In expert-blinded benchmarking on 45 high-complexity cases across six foundation models, HemaGuide substantially improved concordance with tumor board decisions. A systematic ablation study across 11 layers confirmed that performance gains were routing-type-dependent, with no single component sufficient across case types. Automated classification of 70 clinically relevant missense variants showed high concordance with expert standards; no oncogenic variant was downgraded to benign and the whole workflow was completed under real-time conditions on commodity hardware with a median latency of 39 s rather than the hours typically required for manual molecular board workflows. In a simulated practice study, agent-assisted resident physicians achieved near-senior concordance and partially outperformed senior physicians in their subspecialty. External validation on 555 independent cases from a second academic center yielded 81.8% concordance across 47 entities, and a prospective 1-month silent trial on 64 consecutive, unselected cases achieved 82.8% concordance. Hallucinations occurred in 2 of 664 evaluated cases (0.3%). Together these data provide evidence that locally deployable, case-grounded large language model agents can deliver auditable clinical decision support across hematological malignancies, with concordance maintained across institutions and under real-time conditions on commodity hardware.
13 February 2026 in Blood Cancer Discovery
The significant clinical benefit of bispecific T-cell engagers (TCE) for the treatment of relapsed/refractory multiple myeloma (RRMM) may be offset by serious toxicities and treatment failure. Risk scores such as the CAR-HEMATOTOX (HTX), Endothelial Activation and Stress Index (EASIX), and modified EASIX (m-EASIX) can identify patients at risk for complications before chimeric antigen receptor (CAR) T-cell therapy, but their utility prior to TCE therapy remains elusive. We analyzed associations with outcomes and toxicities in independent discovery (n = 123) and validation (n = 155) cohorts treated with TCEs. Patients with HTX ≥3 or m-EASIX > median (>0.86) had a significantly increased risk of prolonged hospitalization, antibiotic treatment, and fever during step-up dosing. We also observed associations with cytopenias requiring therapeutic intervention, higher severe infection and intervention densities, as well as inferior response rates and reduced progression-free and overall survival. Our findings highlight the potential of these clinical scores to improve risk stratification before TCE therapy. Significance: Scores such as HTX, EASIX, and m-EASIX have emerged as helpful tools to enable risk stratification before CAR T-cell therapy. In this study, we demonstrate their utility in patients with RRMM receiving TCEs. HTX ≥3 and m-EASIX > median (>0.86) proved to be risk markers for infections, therapeutic interventions, and poor outcomes. See related commentary by Banerjee and Dhodapkar, p. 345