AUTHORS
Irit Avivi, MD, Miri Neaman, MD2, Chava Perry, MD, PhD3, Erel Joffe, MD4, Roy Vitkon, MD5, Yael Michaeli, PhD6, Noa Gilat7, Lynne Itelson8, Shahar Zirkin9, Jasline Deek, PhD10, Efstathios Kastritis, MD11, Karolis Sablauskas, MD12, Skirmante Cernauskiene13, Vilmante Vaitekenaite14, Andrius Zucenka, MD15, Tomas Papajik, MD, CSc., Prof.16, Eva Kriegova, PhD17, Petr Gajdos18, Yuval Ebenstein, PhD19
AFFILIATIONS
1 Gray Faculty of Medical and Health Sciences, Tel Aviv University, Tel Aviv, Israel, Hematology department, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
2 Department of Chemistry, Raymond and Beverly Sackler Faculty of Exact Sciences, Tel Aviv University, Tel Aviv, Israel, Hematology department, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
3 Gray Faculty of Medical and Health Sciences, Tel Aviv University, Tel Aviv, Israel, Hematology department, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
4 Hematology department, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
5 Hematology department, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
6 Jaxbio Technologies Ltd., Netanya, Israel
7 Jaxbio Technologies Ltd., Netanya, Israel
8 Jaxbio Technologies Ltd., Netanya, Israel
9 Jaxbio Technologies Ltd., Netanya, Israel
10 Department of Chemistry, Raymond and Beverly Sackler Faculty of Exact Sciences, Tel Aviv University, Tel Aviv, Israel
11 Department of Clinical Therapeutics, National and Kapodistrian University of Athens, School of Medicine, Athens, Greece
12 Hematology, Oncology and Transfusion Medicine Center, Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania, Institute of Data Science and Digital Technologies, Vilnius University, Vilnius, Lithuania
13 Hematology, Oncology and Transfusion Medicine Center, Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania
14 Hematology, Oncology and Transfusion Medicine Center, Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania
15 Hematology and Oncology Department, Faculty of Medicine, Vilnius University, Vilnius, Lithuania, Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania
16 Department of Hemato-Oncology, Faculty of Medicine and Dentistry, Palacky University and University Hospital, Olomouc, Czech Republic
17 Department of Immunology, Faculty of Medicine and Dentistry, Palacky University and University Hospital, Olomouc, Czech Republic
18 Department of Computer Science, Faculty of Electrical Engineering and Computer Science, Technical University of Ostrava, Ostrava, Czech Republic
19 Department of Biomedical Engineering, Faculty of Engineering, Tel Aviv University, Tel Aviv, Israel
ABSTRACT
Background: Minimally invasive tools that detect B-cell non-Hodgkin lymphoma (B-NHL) in blood, resolve its subtypes, and track disease status could enable earlier detection, reduce reliance on tissue biopsy and repeated PET/CT, and support longitudinal monitoring. Aberrant DNA methylation is an early, subtype-defining event in lymphomagenesis, yet current methylation profiling depends on sequencing, which is slow, costly, and impractical for routine use. Within the prospective, multicenter SANGUINE study (NCT05735704), we developed a HemaChip, a rapid, sequencing-free epigenetic liquid-biopsy platform that uses DNA extracted from peripheral blood. DNA is chemoenzymatically labeled at methylated CpG sites with fluorescent tags, and a unique epigenetic signature is detected on a targeted microarray, yielding a scalable, low-cost workflow suited to routine laboratories. We hypothesized that DNA methylation signatures could discriminate follicular lymphoma (FL), marginal-zone lymphoma (MZL), and diffuse large B-cell lymphoma (DLBCL) from healthy controls and from one another, and could distinguish active disease from remission, using PET/CT as the reference standard.
Methods: Patients aged >=18 years with newly diagnosed, untreated FL, MZL, or DLBCL (stages I-IV) and age-matched, cancer-free controls were prospectively enrolled. Peripheral blood was collected at diagnosis and after therapy or at progression, and genomic DNA from buffy coat was profiled. In a genome-wide discovery stage, methylation profiles were generated using a high-content array interrogating approximately 1 million genomic loci, and a deep neural network selected 50-200 discriminatory biomarkers for each subtype, both versus healthy controls and versus the other subtypes. The selected biomarkers were then printed on the targeted HemaChip array and evaluated in an independent validation cohort.
Results: Overall, 434 individuals participated in the study, 236 B-NHL patients and 198 controls. In discovery (36 FL, 40 MZL, 45 DLBCL; 56 controls), genome-wide methylation detected each subtype with high accuracy versus controls: FL, sensitivity 88.9%, specifically 94.6%; MZL, 100% and 100%; DLBCL, 91.1% and 98.2%. Signatures further separated indolent lymphomas from aggressive DLBCL, classifying correctly 84.4% of DLBCL versus 88.9% of FL, and 77.8% of DLBCL versus 85% of MZL in respective pairwise comparison. In an initial monitoring set of 17 patients (8 FL, 3 MZL, 6 DLBCL) sampled before and after therapy, all responding samples shifted toward the healthy signature, whereas samples at stable or progressive disease retained the disease signature, concordant with PET/CT.
The targeted HemaChip was validated with a cohort of 349 participants (192 B-NHL, stages I-IV; 157 controls; 58.9% female; median age 67 years [IQR 53-75]), closely reproducing the discovery results. Preliminary HemaChip performance was FL- sensitivity 81.2%, specificity 95.2%, AUC 0.904; MZL- 87.2%, 89.8%, AUC 0.92, and DLBCL- 95.1%, 94.3%, AUC 0.98.
Conclusions: HemaChip, a simple, sequencing-free, epigenetic blood test, accurately detects and subtypes B-cell lymphomas directly from peripheral blood. The successful translation of genome-wide methylation biomarkers into a rapid, targeted assay, together with early concordance with PET/CT for disease status, supports HemaChip as a practical, scalable, minimally invasive tool for lymphoma detection, subtyping, treatment-response assessment, and longitudinal surveillance.
DISCLOSURES
Irit Avivi, MD: Roche - Consultancy (Includes expert testimony); Johnson & Johnson - Consultancy (Includes expert testimony); AbbVie - Consultancy (Includes expert testimony)