Recommendations for using artificial intelligence in clinical flow cytometry.

Flow cytometry is a key clinical tool in the diagnosis of many hematologic malignancies and traditionally requires close inspection of digital data by hematopathologists with expert domain knowledge. Advances in artificial intelligence (AI) are transferable to flow cytometry and have the potential to improve efficiency and prioritization of cases, reduce errors, and highlight fundamental, previously unrecognized associations with underlying biological processes. As a multidisciplinary group of stakeholders, we review a range of critical considerations for appropriately applying AI to clinical flow cytometry, including use case identification, low and high risk use cases, validation, revalidation, computational considerations, and the present regulatory frameworks surrounding AI in clinical medicine. In particular, we provide practical guidance for the development, implementation, and suggestions for potential regulation of AI-based methods in the clinical flow cytometry laboratory. We expect these recommendations to be a helpful initial framework of reference, which will also require additional updates as the field matures.

An Integrated Analysis of Heterogeneous Drug Responses in Acute Myeloid Leukemia That Enables the Discovery of Predictive Biomarkers.

Many promising new cancer drugs proceed through preclinical testing and early-phase trials only to fail in late-stage clinical testing. Thus, improved models that better predict survival outcomes and enable the development of biomarkers are needed to identify patients most likely to respond to and benefit from therapy. Here, we describe a comprehensive approach in which we incorporated biobanking, xenografting, and multiplexed phospho-flow (PF) cytometric profiling to study drug response and identify predictive biomarkers in acute myeloid leukemia (AML) patients. To test the efficacy of our approach, we evaluated the investigational JAK2 inhibitor fedratinib (FED) in 64 patient samples. FED robustly reduced leukemia in mouse xenograft models in 59% of cases and was also effective in limiting the protumorigenic activity of leukemia stem cells as shown by serial transplantation assays. In parallel, PF profiling identified FED-mediated reduction in phospho-STAT5 (pSTAT5) levels as a predictive biomarker of in vivo drug response with high specificity (92%) and strong positive predictive value (93%). Unexpectedly, another JAK inhibitor, ruxolitinib (RUX), was ineffective in 8 of 10 FED-responsive samples. Notably, this outcome could be predicted by the status of pSTAT5 signaling, which was unaffected by RUX treatment. Consistent with this observed discrepancy, PF analysis revealed that FED exerted its effects through multiple JAK2-independent mechanisms. Collectively, this work establishes an integrated approach for testing novel anticancer agents that captures the inherent variability of response caused by disease heterogeneity and in parallel, facilitates the identification of predictive biomarkers that can help stratify patients into appropriate clinical trials.

IL-7 coordinates proliferation, differentiation and Tcra recombination during thymocyte ß-selection.

Signaling via the pre-T cell antigen receptor (pre-TCR) and the receptor Notch1 induces transient self-renewal (ß-selection) of TCRß(+) CD4(-)CD8(-) double-negative stage 3 (DN3) and DN4 progenitor cells that differentiate into CD4(+)CD8(+) double-positive (DP) thymocytes, which then rearrange the locus encoding the TCR α-chain (Tcra). Interleukin 7 (IL-7) promotes the survival of TCRß(-) DN thymocytes by inducing expression of the pro-survival molecule Bcl-2, but the functions of IL-7 during ß-selection have remained unclear. Here we found that IL-7 signaled TCRß(+) DN3 and DN4 thymocytes to upregulate genes encoding molecules involved in cell growth and repressed the gene encoding the transcriptional repressor Bcl-6. Accordingly, IL-7-deficient DN4 cells lacked trophic receptors and did not proliferate but rearranged Tcra prematurely and differentiated rapidly. Deletion of Bcl6 partially restored the self-renewal of DN4 cells in the absence of IL-7, but overexpression of BCL2 did not. Thus, IL-7 critically acts cooperatively with signaling via the pre-TCR and Notch1 to coordinate proliferation, differentiation and Tcra recombination during ß-selection.