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

Paper Details

Abstracts, authors, affiliations, and disclosures.

SESSION

655. Multiple Myeloma: Cellular Therapies: Poster III

Program: Poster Abstract Session
Monday, December 14, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Omics Technologies, Bioinformatics, Cell Expansion, Computational Biology, Treatment Considerations, Combination Therapy, Drug Development, Therapy Sequence, Biological Therapies, Bispecific Antibody Therapy, Chimeric Antigen Receptor, (CAR)-T Cell Therapies, Immunotherapy, Adverse Events, Artificial Intelligence (AI), Deep Learning, Generative AI, Machine Learning, Natural Language Processing, Biological Processes, Microbiome, Multi-Systemic Interactions, Clinical Practice (Health Services and Quality), Systems-Based Hematology, Diseases, Lymphoid Malignancies, Plasma Cell Disorders, Myeloid Malignancies, Research, Translational Research, Clinical Research, Clinical Trials, Real-World Evidence, Registries, Study Population, Human, Adult
PAPER

5796 Longitudinal serum and CSF proteomics links TFF3, OPG, and chemokine signaling to CART–Associated immune-related adverse events in multiple myeloma

Luca Paruzzo, MD, Matthew Ho, MBChB2, Patrizia Porazzi, PhD3, Peter Michener4, Alberto Carturan, MD5, Adam Waxman, MD, MS6, Shivani Kapur, MD7, Anushka Anant Padmanabhan8, Priyamvada Das9, Rohith Raali10, Sandra Susanibar-Adaniya, MD, MS11, Edward A. Stadtmauer, MD12, Dan T. Vogl, MD, MSCE13, Caitlin Hansen14, Jen Gillies15, Kerry Scott16, Jacob N. Wohlstadter17, Hans Layman18, Alfred L. Garfall, MD19, Adam D. Cohen, MD20, Joseph Fraietta, PhD21, Marco Ruella, MD22
1 Lymphoma Program, Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, United States
2 Center for Cellular Immunotherapies, University of Pennsylvania, Philadelphia, Pennsylvania, United States
3 Lymphoma Program, Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, United States
4 Center for Cellular Immunotherapies, University of Pennsylvania, Philadelphia, Pennsylvania, United States
5 Lymphoma Program, Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, United States
6 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
7 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
8 Center for Cellular Immunotherapies, University of Pennsylvania, Philadelphia, Pennsylvania, United States
9 Center for Cellular Immunotherapies, University of Pennsylvania, Philadelphia, Pennsylvania, United States
10 Center for Cellular Immunotherapies, University of Pennsylvania, Philadelphia, Pennsylvania, United States
11 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
12 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
13 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
14 Meso Scale Diagnostics, LLC, Rockville, Maryland, United States
15 Meso Scale Diagnostics, LLC, Rockville, Maryland, United States
16 Meso Scale Diagnostics, LLC, Rockville, Maryland, United States
17 Meso Scale Diagnostics, LLC, Rockville, Maryland, United States
18 Meso Scale Diagnostics, LLC, Rockville, Maryland, United States
19 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
20 Division of Hematology-Oncology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
21 Center for Cellular Immunotherapies, University of Pennsylvania, Philadelphia, Pennsylvania, United States
22 Lymphoma Program, Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, United States

Background: Serum captures both tumor- and host-derived signals, alongside the systemic immune and inflammatory state. Because CAR T cells expand and traffic within this systemic environment, soluble proteins may influence both efficacy and toxicity. Ciltacabtagene autoleucel (cilta-cel) produces durable remissions in patients with multiple myeloma (MM) but is associated with delayed complications collectively termed CAR T-cell–associated immune-related adverse events (CirAEs, Ho et al, Nat Med 2026). We hypothesized that longitudinal serum proteomics could identify biological programs and soluble predictors and mediators of CirAEs.

Methods: We performed longitudinal multiplex proteomic profiling with absolute quantitation using the ProteinXI Discovery 400 high multiplex immunoassay in 133 serum and 7 CSF samples (5 neuro-CirAE and 2 controls) from 47 MM patients treated with cilta-cel (April 2022-July 2025). All CirAE patients with samples available in our center were included, together with selected comparator patients without CirAE. Serum samples were collected before lymphodepletion (pre-LD, n=41), and on day 7 (n=45), 14 (n=32), and 28 (n=15) post-cilta-cel. Principal component analysis (PCA) was used to assess the separation of samples by clinical outcomes and collection time points. Differential protein-expression (DEP) analyses were performed across longitudinal time points and between CirAE groups.

Results: The analytic cohort included 23 patients with CirAE and 24 without. CirAE types were: neuro-CirAEs (n=16), enterocolitis (n=4), inflammatory arthritis (n=2), and multiple CirAEs in the same patient (facial palsy, enterocolitis, n=1)

Cilta-cel induced a time-dependent shift in the circulating proteome from early systemic inflammation to later T-cell activation/exhaustion. Overall, 76 (19%) of the proteins significantly changed after cilta-cel infusion. PCA demonstrated separation of pre-LD samples from days 7 and 14. The day 7 proteomic signature was dominated by inflammatory cytokines and hematopoietic growth factors, including IL-3, IL-5, CCL1, IL-15, and G-CSF. By day 14, a distinct T-cell activation/exhaustion-associated signature emerged, characterized by increased soluble GZMB, CD27, CD28, TACTILE, TIGIT, and PD-1. In contrast, soluble TACI and BCMA concentrations were highest at baseline and declined as early as day 7.

The pre-LD serum proteome distinguished patients who subsequently developed CirAEs(n=17) from those who did not (n=24). PCA demonstrated group-level separation between the two groups. Analysis of the proteins contributing most strongly to this separation identified trefoil factor 3 (TFF3) and osteoprotegerin (OPG) as significantly lower in patients with CirAEs (FDR=0.04 for both). These associations were corroborated by DEP analysis across the complete protein panel. Notably, TFF3 has been reported to suppress microglial production of inflammatory cytokines and to exert protective effects in intestinal inflammation.

Patients who developed CirAEs subsequently demonstrated a more pronounced early inflammatory response to cilta-cel. From pre-LD to day 7, patients who developed CirAEs exhibited significantly greater increases in multiple chemokines and inflammatory mediators, including MCP-2 (FDR=0.01), hepcidin (FDR=0.02), TNF-α (FDR=0.02), and IL-6 (FDR=0.02). MCP-2 is a ligand for CCR5, a signaling axis previously implicated in lymphocyte expansion associated with CirAEs.

Exploratory CSF profiling identified a shared proteomic pattern among patients with neuro-CirAEs. PCA of samples from 5 patients with neuro-CirAEs and 2 controls demonstrated clustering of neuro-CirAE samples, suggesting a shared proteomic profile. Angiopoietin-4 and OPG contributed strongly to this separation and were higher in neuro-CirAE CSF (both P<0.05).Finally, we compared the CSF and matched serum proteome to identify chemotactic drivers recruiting CART in the CSF. Across 15 chemokines, only CXCL16 and MCP1 were enriched in CSF, suggesting a potential role for recruiting immune cells in the CirAE CSF.

Conclusion: We identified low pre-LD levels of TFF3 and OPG as associated with CirAE. Compared with baseline, CirAE patients showed increased inflammatory cytokines, with MCP-2 among the top hits supporting the mechanistic role of the CCR5 axis. In neuro-CirAE, CXCL16 and MCP-1 may promote immune-cell recruitment into the CSF. These pathways warrant mechanistic evaluation as potential targets for CirAE prevention.

 

Luca Paruzzo, MD: University of Pennsylvania - Patents & Royalties