2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Monday, December 14, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Assays, Imaging, Molecular Testing, Newborn Screening, Omics Technologies, Pathology, Serologic Tests, Bioinformatics, Computational Biology, Emerging Technologies, Profiling, Treatment Considerations, Combination Therapy, Therapy Sequence, Biological Therapies, Antibody Therapy, Transplantation, Bispecific Antibody Therapy, (CAR)-T Cell Therapies, Monoclonal Antibody Therapy, Non-Biological Therapies, Chemotherapy, Surgical, Adverse Events, Artificial Intelligence (AI), Computer Vision, Deep Learning, Generative AI, Machine Learning, Natural Language Processing, Biological Processes, Pathogenesis, CHIP, Genomics, Hematopoiesis, Immune Mechanism, Metabolism, Molecular Biology, Multi-Systemic Interactions, Clinical Practice (Health Services and Quality), Clinical Guidelines, Diseases, Hemoglobinopathies, Sickle Cell Disease, Immunology, GVHD, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, Measurable Residual Disease, ALL, CLL, Non-Hodgkin Lymphoma, B Cell Lymphoma, Indolent Lymphoma, Aggressive Lymphoma, Research, Fundamental Science, AML, APL, Translational Research, MDS, MPN, CMML, Clinical Research, Clinical Trials, Health Outcomes Research, Patient-Reported Outcomes, Real-World Evidence, Registries, Survivorship, Study Population, Human, Adult, Elderly, Pediatric
PAPER

6104 Predicting progression across the monoclonal gammopathy-smoldering multiple myeloma-active multiple myeloma continuum using nanophotonic rseira-FTIR and machine learning analysis of serum

David Dingli, MD, PhD, Eng Hock Lee, PhD2, Darlyna Khonkhammy3, Saemundur Rögnvaldsson4, Sigrún Thorsteinsdottir5, Thorvardur J Love6, Björn Leví Gunnarsson7, Sigurdur Kristinsson, MD, PhD8, Dan Que Pham9, Ali Khammanivong, PhD, MS10
1 Mayo Clinic, Rochester, Minnesota, United States
2 Oncodea Corporation, Saint Paul, Minnesota, United States
3 Oncodea Corporation, Saint Paul, Minnesota, United States
4 University of Iceland, Reykjavík, Iceland
5 University of Iceland, Reykjavík, Iceland
6 University of Iceland, Reykjavík, Iceland
7 University of Iceland, Reykjavík, Iceland
8 University of Iceland, Reykjavík, Iceland
9 Oncodea Corporation, Saint Paul, Minnesota, United States
10 Oncodea Corporation, Saint Paul, Minnesota, United States

Monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM), and active multiple myeloma (MM) represent a continuum across the spectrum of plasma cell disorders. MGUS and SMM carry a variable risk of progression to MM, with SMM having the higher risk. Determining this risk of progression is important for proper patient care, particularly given the availability of approved therapies for SMM: patients at low risk of progression may be spared unnecessary intervention, while identifying those at high risk may reduce morbidity through more active surveillance or the timely introduction of therapy. Current risk models use clinical and laboratory parameters to stratify patients into broad risk categories, but substantial uncertainty remains for individual patients, particularly those at intermediate risk.  

We hypothesized that a nanophotonic molecular-sensing approach, combining resonant surface-enhanced infrared absorption with Fourier-transform infrared spectroscopy of serum and machine learning (rSEIRA-FTIR/ML), could support this decision-making by stratifying patients with MGUS and SMM according to their estimated risk of progression to MM. Our prior studies have shown that this is feasible, and here we validate our approach using an independent, screen-detected cohort from the nationwide iStopMM (Iceland Screens, Treats, or Prevents Multiple Myeloma) study.  

Methods: After Institutional Review Board approval, we identified 51 patients with MGUS who had a baseline serum sample available as well as follow-up to ascertain outcomes. Serum molecular fingerprints were then analyzed using a nanophotonic rSEIRA-FTIR platform coupled with an ML model previously trained on a curated Mayo Clinic serum cohort to identify SMM patients who will progress to MM within two years post-biopsy (ASH 2024). Model predictions were compared with independently ascertained clinical outcomes. 

Results: All 51 patients had MGUS at initial diagnosis. Median age was 70 years (range, 43-88), and 24 (47%) were female. Over a follow-up of up to 9 years, 24 of the 51 patients progressed to SMM, of whom 5 further progressed to MM. The median serum M-spike was 1.4 g/L and 3 g/L for MGUS and SMM, respectively. IMWG progression risk score at baseline was 0/1/2 in 33%/43%/24% of all 51 patients and in 29%/38%/33% of the 24 who progressed to SMM. The median follow-up for the cohort was 6.94 years (range, 2-8.7 years). Overall, the ML model correctly predicted progression risk in 45 of 51 patients (88.2% accuracy) within 4 years after initial diagnosis, corresponding to a specificity of 88% (44/50). Of the five MM events, the one occurring within 2 years was correctly predicted as progressing. The other four occurred at 4.4, 5.8, 7.3, and 8.0 years post-biopsy and were correctly predicted as non-progressing within 4 years. One of these four was subsequently diagnosed as SMM seven months after the MM diagnosis and remained SMM at last follow-up (8.7 years). By IMWG risk group, 4-year accuracy was 11/12 (92%) for group 2, 20/22 (91%) for group 1, and 14/17 (82%) for group 0. 

Conclusion: In this independent cohort, a single serum test obtained at diagnosis predicted progression from MGUS or SMM to active MM with high accuracy. The model was trained to predict progression within 2 years, yet accuracy remained substantial over a median follow-up of 6.94 years, suggesting that predictive performance may extend beyond the training window. These findings support nanophotonic rSEIRA-FTIR coupled with machine learning as a minimally invasive approach to progression-risk assessment in patients with MGUS or SMM and warrant evaluation in larger prospective cohorts to determine its potential clinical utility alongside the current standard of care. 

David Dingli, MD, PhD: Sorrento - Honoraria; Regeneron - Honoraria; MSD - Honoraria; Alexion - Honoraria; Sanofi - Honoraria; Janssen - Honoraria; Bristol Myers Squibb - Honoraria; Novartis - Honoraria; Apellis - Honoraria; K36 Therapeutics - Research Funding; Novartis - Consultancy (Includes expert testimony); Sorrento - Consultancy (Includes expert testimony); Regeneron - Consultancy (Includes expert testimony); Sanofi - Consultancy (Includes expert testimony); Genentech - Consultancy (Includes expert testimony); Apellis - Consultancy (Includes expert testimony); MSD - Consultancy (Includes expert testimony); Janssen - Consultancy (Includes expert testimony); Alexion - Consultancy (Includes expert testimony); Bristol Myers Squibb - Consultancy (Includes expert testimony)