Background and significance Indications for cytopenia range from simple reversible causes to serious disorders like myelodysplastic syndromes (MDS). MDS diagnosis currently relies on invasive bone marrow (BM) sampling, a procedure that is uncomfortable, difficult to repeat, often inconclusive, requires trained medical personnel, hospital resources, and is subject to inter-observer variations. This highlights the need for accurate, accessible, and non-invasive diagnostic alternatives.

To address this unmet need, we developed a diagnostic pipeline based on single-cell RNA sequencing (scRNA-seq) of circulating hematopoietic stem and progenitor cells (cHSPCs) from peripheral blood (PB). This test outputs the full information required for MDS diagnosis from a simple blood test, including cytogenetics, mutations, blast frequency, and cellular composition. We recently profiled cHSPCs from PB of 148 healthy individuals (Furer et al., Nature Medicine 2025) to generate a reference single-cell transcriptomics atlas. As a proof-of-concept diagnostic tool, this atlas was used to create a computational model that distinguishes MDS from cytopenia. Applying this model on 73 individuals with unexplained cytopenia or confirmed MDS resulted in high specificity and sensitivity (AUC 0.93 and 0.97 on the train and test sets, respectively). This simple and non-invasive test is more cost-effective than BM procedures, enables frequent monitoring in a watchful-waiting regime, and offers the potential for high accuracy via an automated analysis pipeline.

In this clinical trial, we propose a prospective, multi-center clinical trial aimed at validating the diagnostic pipeline. Specifically, we aim to assess whether our approach can reliably predict bone marrow pathology for patients referred to BM examination due to unexplained cytopenia and ultimately eliminate the need for BM biopsies.

Study design and methods PERIBLOOD study (NCT07081087) aims to extend and validate our diagnostic pipeline across diverse geographic and ethnic populations. Participating centers in Europe, the United States, South America, Asia, and Africa are currently enrolling participants, with sample collection occurring both prospectively (from hospitals) and retrospectively (from biobanks), expecting to enroll 1500 participants within 3 years.

Eligible participants include individuals with cytopenia or suspected cytopenia referred for BM examination. Inclusion criteria require at least one of the following: platelet count <150 × 10⁹/L, absolute neutrophil count <1.8 × 10⁹/L or hemoglobin <13 g/dL (males) / <12 g/dL (females), in the absence of iron, folate, or vitamin B12 deficiency. Participants with leukemia or those undergoing disease-related therapy are excluded.

For each enrolled participant, 45 mL of PB will be collected within ±2 months of the BM sampling date. Peripheral blood mononuclear cells (PBMCs) will be isolated via ficoll extraction, frozen, and shipped to the Weizmann Institute of Science for processing, together with the BM analysis and the complete blood count (CBC) results. PBMCs from 5-10 samples will be pooled together, undergo CD34 enrichment for cHSPCs, and profiled with scRNA-seq. Each sample will participate in 2 pooled runs, and each run will include a control sample from a healthy individual, to ensure robustness and batch effect handling. Additional assays include 2 targeted sequencing panels: a mutation panel covering key myeloid malignancy genes, and a genotyping panel to enable demultiplexing of the pooled scRNA-seq samples.

The final diagnostic reports will include the PB assay test results with the matched BM results, and will be sent to the treating physicians for clinical interpretation and treatment planning based on the integrated findings. The report will include an estimation of the BM blast frequency, karyotype information, mutations in key genes, and additional novel markers, all derived from the PB sample, which will be integrated into a final diagnosis. To evaluate the computational model we will compare the BM results obtained in parallel to our PB test. A random set of 100 samples will be sent to an additional hemato-pathologist for re-analysis to assess the stability of the BM-based diagnosis in conjunction with our PB-based one.

This is the world's first PB single-cell test to replace BM aspiration for hematological diagnostics and monitoring, and it will eventually expand to become a comprehensive hematologic diagnostics platform.

This content is only available as a PDF.
2025
Sign in via your Institution